Note: Descriptions are shown in the official language in which they were submitted.
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Improved Method of Mapping Glycans of Glycoproteins in Serum Samples
Description
Field of the Invention
The present invention relates to a method for analyzing glycans of a
recombinant glycoprotein in a
liquid sample of a mammal. Specifically the method comprises a step of
affinity purifying the
recombinant glycoprotein from the sample, enzymatically releasing a glycan
containing fragment from
the immobilized glycoprotein, adding a reference standard containing
isotopically labeled glycans,
fluorescently label the glycans and analyzing the glycans using LC-MS. The
present invention also
relates to a method further comprising analyzing the glycans of the
immobilized recombinant
glycoprotein fragment, further comprising a pre-clearing step of the liquid
sample, and releasing the
glycans from the immobilized recombinant glycoprotein fragment. This method is
particularly suited to
measure pharmacokinetic parameters of a recombinant glycoprotein, such as a
biopharmaceutical, in a
mammal in clinical or pre-clinical studies. The present invention may be used
with liquid
chromatography-mass spectrometry, preferably reverse phase nano-liquid
chromatography-mass
spectrometry (RP-nano-LC-MS) technology. This allows for the use of a small
sample volume and the
possibility to operate with high throughput, such as in a 96-well plate sample
preparation. It also
provides high sensitivity and allows analysis of glycan structures
individually.
Background of the Invention
Biological drugs belong to the most complex pharmaceuticals. One important
class of biological drugs
are glycoproteins, and in particular monoclonal antibodies (mAbs), which are
typically glycoproteins with
a molecular mass of about 150 kDa. In recent years the glycosylation of
therapeutic proteins came into
focus as several investigations showed that these structures may have an
effect on the safety and
efficacy of the corresponding biological drugs. Further, possible effects of
glycan structures on the
pharmacokinetics (PK) have become of interest.
IgGs have a conserved N-glycosylation site on each heavy chain at their Fc
part, usually around amino
acid 297. The heterogeneity of the N-glycans attached to these sites is very
high and more than a
dozen different glycans can be found. Identification and quantification of
glycans from this mixture is
sophisticated and demands a comprehensive analytical approach. Structural
isomers of several glycans
make the discrimination even more difficult.
N-glycosylation is a critical issue for the safety and efficacy of
biopharmaceuticals and has a possible
influence on the pharmacokinetic of a glycoprotein, such as for therapeutic
antibody or therapeutic Fc-
fusion proteins. Hence, comprehensive analysis of individual glycan structures
of a glycoprotein on the
1
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
pharmacokinetics is necessary. The interest in Fc glycans involves the
following major topics, the role
of Fc glycans in antibody-dependent cell-mediated cytotoxicity (ADCC) and
complement dependent
cytotoxicity (CDC) and the role of glycans on clearance rate from the
circulation as a result of
glycoreceptors such as the asialoglycoprotein receptor that binds terminal
galactose residues or
mannose receptors that bind terminal mannose and N-acetylglucosamine residues.
N-glycosylation can be studied after enzymatic release of the glycans by
PNGaseF using liquid
chromatography, mass spectrometry or by a combination of both technologies
(Chen and Flynn,
Analytical Biochemistry 2007, 370, 147-61). To increase sensitivity, the N-
glycans are often derivatized
with fluorescence labels. For quantitative analytics of N-glycans, labeling
with a fluorescence dye
through reductive amination is widely used, offering high sensitivity. Every N-
glycan carries only one
label independent of size or branching, therefore, quantitative information
can be obtained from the
fluorescence signal intensity. The tag can not only provide UV or fluorescence
detection but also the
efficiency in electrospray ionization can be improved. For the analysis of
sialic acid-containing glycans,
labeling with a negatively charged tag, like 2-aminobenzoic acid (2-AA) or 8-
aminonaphthalene-1,3,6-
trisulfonic acid (ANTS), and MS detection in negative ionization mode is
frequently used (Prien et al.,
Analytical Chemistry 2010, 82, 1498-508).
Negatively charged labeling, in combination with negative ionization in MS,
has been frequently used to
analyze acidic N-glycans (Prien et al., Glycobiology 2010, 20, 629-47). For
the characterization of
oligomannose structures, 2-AA derivatized N-glycans were analyzed by a
combination of reversed
phase chromatography and negative mode mass spectrometry (Prien et al.,
Glycobiology 2010, 20,
629-47).
In WO 2014/060551, an efficient and sensitive analysis of neutral and acidic N-
glycans by RP-LC-MS in
one approach is described. Previous methods have been mostly limited to the
isomer separation of
oligomannose type N-glycans. In WO 2014/060551 anthranilic acid (2-AA) is used
to label N-glycans
prior to separation using a reversed-phase liquid chromatography (RP-LC)
column under acidic
conditions using formic acid. Negatively charged 2-AA offers stronger
retention on the reversed phase
column than 2-aminobenzamide (2-AB) in RP-LC and allows efficient ionization
and detection of 2-AA
labeled N-glycans. The acidic conditions used for the RP-LC lead to an
efficient separation of acidic 2-
AA N-glycans carrying terminal sialylation without the need for an ion-pairing
reagent. This method may
be used with RP-nano-LC-MS and a 96-well plate sample preparation, which
allows attomolar
sensitivity and high throughput.
N-glycosylation analysis is sophisticated because of countless N-glycan
variants which may be
attached to the protein molecules and the huge differences in their relative
amounts. For example, for
recombinant human IgG antibodies relative amounts of individual oligomannose N-
glycans may range
from 0.02% to more than 70% for the most abundant N-glycan, differences that
cover three orders of
magnitude (Nigel et al., Analytical and Bioanalytical Chemistry 2013, 405,
2481-93). Technologies
frequently used for N-glycan analysis are CE, HPAEC-PAD, HPLC, MALDI and ESI-
MS and various
2
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
combinations of these technologies (Marino etal., Nature Chemical Biology
2010, 6,713-23). LC-MS is
an advantageous combination as LC is able to separate glycan mixtures;
afterwards glycan variants
can then individually be identified and quantified by on-line MS. For various
analytical applications,
however, conventional LC-MS may not be sufficiently sensitive, especially for
cases where the sample
amount is strongly limited. During early biopharmaceutical development (clone
or pool selection) and
during pre-clinical studies, only minute amounts of recombinant protein from
micro titer plates are
usually available for protein and glycan analysis. For studying the influence
of glycan variants on the
pharmacokinetics of glycosylated biopharmaceuticals, the recombinant
glycoprotein of interest is further
present in a sample, e.g., serum, in admixture with other proteins, most of
which are also glycosylated.
Further, the amount of the recombinant glycoprotein of interest declines over
time. Thus, high
sensitivity, specificity and low sample consumption is required. Typically
there is a large sample
number, hence the glycoprotein of interest must be isolated from the sample in
a reproducible manner
ideally in a high throughput format and the analysis of the individual samples
within a series of samples
must be comparable to each other. Moreover, the glycoprotein may have more
than one N-
glycosylation site, with differences in glycan structures at each
glycosylation site.
In proteomics, limitations in sample volume have been circumvented by reducing
the dimensions of the
analytical system, for example, by the use of nano-LC-MS. Approaches for N-
glycan analysis by the
use of nano-LC-MS are rare in the literature. Three investigations have
reported feasibility of nano-LC
for glycan analysis. Wuhrer et al. have miniaturized HILIC-MS to nanoscale for
oligosaccharide analysis
(Wuhrer et al., Analytical Chemistry 2004, 76, 833-8). They have analyzed
underivatized N-glycans
with femtomolar sensitivity. Avoiding glycan derivatization shortens sample
preparation but the benefit
of improved MS detection coming with the label is lost (Wuhrer et al.,
Analytical Chemistry 2004, 76,
833-8; Ruhaak et al., Analytical and Bioanalytical Chemistry 2010, 397, 3457-
81). Kalay et al. have
used normal phase nanoscale HPLC-MS with on-line fluorescence to analyze 2-AB
N-glycans
(Analytical Biochemistry 2012, 423, 153-162). Their approach resulted in long
and time consuming
gradients to achieve a good chromatographic resolution. One work utilizing
nano-reversed phase
chromatography (nano-RPC) has been published recently. Ritamo et al.
(Analytical and Bioanalytical
Chemistry 2013, 405, 2469-80) have used a nano-LC system to separate
permethylated N-glycans.
They have achieved separation for various structural isomers on a nano-RP-
column. Permethylation of
N-glycans, however, is complex and toxic reagents are used and formation of
side products is rather
likely which makes routine use questionable. It has been reported previously
that RP-LC with on-line
MS is broadly applicable for analysis of differently reducing-end labeled N-
glycans (Chen and Flynn,
Analytical Biochemistry 2007, 370, 147-61; Prater et al., Analytical
Biochemistry 2009, 385, 69-79). An
overall sensitivity for a single 2-AA labeled N-glycan, e.g., about 400
attomole using RP-nano-LC-MS
was reported in WO 2014/060551. This method may further be combined with a 96
well based sample
preparation work-flow.
Quantification of proteins by MS is frequently performed using stable heavy
isotope peptides or
proteins. Stable heavy isotope labeling of N-glycans in contrast is hardly
possible due to the complex
3
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
and multiple connected anabolic and catabolic carbohydrate pathways of
eukaryotic cells. The
controlled incorporation of isotopically labeled monosaccharides is far more
challenging than the
incorporation of isotopically labeled amino acids like lysine or arginine.
The role of glycans in clearing IgG molecules in circulation has been examined
by various groups,
including pre-clinical or clinical pharmacokinetic (PK) studies, in an attempt
to determine the impact of
Fc glycan changes on serum clearance. Two general approaches have been used in
these studies. In
one approach, antibodies have been prepared that are enriched in specific
glycan forms, either through
genetic mutation, addition of metabolic inhibitors, a combination of both,
affinity purification or
enzymatic treatment. These antibody samples are then injected into animals to
determine the impact on
the overall clearance rate. However, this approach has limitations because of
the assumption that the
only change in the molecule during enrichment is the glycan structure. Glycan
heterogeneity can also
lead to some ambiguity of the results. Also different variants of the same
product need to be prepared
and administered. Thus, this approach is not suitable for pre-clinical or
clinical pK studies of
biopharmaceuticals, such as therapeutic recombinant antibodies or fusion
proteins.
In another approach, the glycan forms are analyzed after the drug has been
administered and changes
to the glycan pattern with circulation time are interpreted as differences in
clearance rates. The post-
administration collection approach can follow many more glycan forms
simultaneously. Since changes
to a single administered sample are followed, the impact of specific
microheterogeneity can be
monitored. To the best of our knowledge this approach has only been used in
humans, but not in
smaller mammals typically used in pre-clinical studies, where sample size is
more limiting. However,
analysis of glycans on pharmacokinetics as early as possible in the
development of biopharmaceuticals
is desirable, as well as reducing the sample size needed in clinical trials.
Alessandri et al (mAbs, 2012, 4(4), 509-520) analyzed a recombinant monoclonal
IgG1 antibody from
serum samples obtained from a human PK study. The antibody was purified from
serum by affinity
chromatography using its ligand cross-linked to sepharose beads. The glyans
were released from the
acidically eluted antibody, labeled with 2-aminobenzamide (2-AB) and analyzed
by normal phase high
performance liquid chromatography. However, the assay allowed analysis only up
to 14 days post-
administration in a rather high sample volume and is prone to contamination
with other serum proteins.
Chen et al (Glycobiology 2009, 19(3), 240-249) analyzed monoclonal antibodies
in serum samples of a
pharmacokinetic study in humans. The antibodies were affinity purified using a
ligand cross-linked to a
resin and acidically eluted thereof. The released glycans were labeled with 2-
aminobenzamide (2-AB)
and analyzed by RP-HPLC/MS. This isolation method was capable of obtaining
only 70% of the specific
antibody from the serum. Also, this method is subject to interference from
glycans released from other
serum proteins or other glycans of the same molecule for glycoproteins with
more than one
glycosylation sites.
Goetze et al (Glycobiology 2011, 21(7), 949-959) analyzed monoclonal IgG1 and
IgG2 antibodies in
serum samples of a pharmacokinetic study in humans. The antibodies were
affinity purified using the
4
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
respective ligand cross-linked to a resin. The antibodies were acidically
eluted and digested with either
the endoproteinase Lys-C or trypsin. The glycosylated peptides were analyzed
using LC-MS/MS. This
method offers the advantage of being specific for the consensus Fc
glycosylation of either human IgG1
or human IgG2. The applied peptide mapping approach reduces the potential pool
of interfering
impurities to only endogenous IgG of the same subclass as the mAb to be
analyzed. However, this
method requires a sample volume of 0.5 ml, which often may not be available,
e.g., in preclinical
studies using rodents. Also this analysis allows only for determining the
relative proportion of N-glycan
structures within a sample, but not for an independent analysis of single N-
glycan structures.
None of these methods analyzed recombinant glycoproteins other than antibodies
that may comprise
more than one glycosylation site with structural differences at the
glycosylation sites, such as in the two
domains of a fusion protein. Further, there was a need for a method to analyze
individual glycans of a
recombinant glycoprotein in a series of mammalian samples, such as serum,
rather than a proportion of
the glycans within the sample. There was also a need for high sensitivity to
allow for low sample
consumption and a sample preparation to be operated in a high throughput
manner.
Summary of the Invention
Accordingly, the present invention relates to a method of analyzing glycans of
a recombinant
glycoprotein of interest in liquid samples of a mammal comprising a) providing
two or more liquid
samples of a mammal comprising the recombinant glycoprotein of interest; b)
immobilizing the
recombinant glycoprotein of each of said samples on a separate solid support
coupled to an affinity
ligand specific for the recombinant glycoprotein in the samples; c) releasing
a glycan containing
fragment of the recombinant glycoprotein of each of said samples from the
solid support into separate
eluates by enzymatic cleavage of the recombinant glycoprotein; d) optionally
releasing the glycans from
the glycan containing fragment of the recombinant glycoprotein of each of said
samples in the separate
eluates; e) labeling the glycans of each of said samples with a first stable
isotope of a fluorescent label;
and f) analyzing the labeled glycans of step e) of each of said samples
separately using LC-MS and
comparing said two or more samples, wherein a reference standard comprising
glycans labeled with a
second stable isotope of the fluorescent label is added to each of said
samples prior to step c), d), e) or
f) and the reference standard is analyzed together with the labeled glycans of
step f).
The present invention also relates to a method further comprising analyzing
the glycans of a second
glycan containing fragment of the recombinant glycoprotein of each of said
samples that remains
immobilized to the solid support in step c), comprising the following steps:
aa) pre-clearing each of the
two or more liquid samples of a mammal to remove glycoproteins binding non-
specifically to the solid
support prior to step b); dd) releasing the glycans from the immobilized
second glycan containing
fragment of the recombinant glycoprotein of each of said samples remaining on
the solid support
following step c) into separate solutions; ee) labeling the glycans of each of
said samples with a first
stable isotope of a fluorescent label; and ff) analyzing the labeled glycans
of step ee) of each of said
5
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
samples separately using LC-MS and comparing said two or more samples, wherein
a reference
standard comprising glycans labeled with a second stable isotope of the
fluorescent label is added to
each of said samples prior to step dd), ee) or if) and wherein the reference
standard is analyzed
together with the labeled glycans of step if). The pre-clearing step aa) may
comprise contacting the
samples with a solid support, wherein the solid support is made of the same
material as the solid
support used in step b), but is not coupled to said affinity ligand.
According to the present invention the glycans to be analyzed may be N-glycans
selected from the
group consisting of high mannose type, hybrid type or complex type N-glycans.
The fluorescent label as used in the methods of the invention has a stable
light and a stable heavy
isotope variant and may be for example 2-amino benzamide (2-AB) or 2-amino
benzoic acid (2-AA).
Preferably, the fluorescent label is a 12[C] and 13[C] isotopic pair, more
preferably a 12[C6] and 13[C6]
isotopic pair, and even more preferably a 12[C6]-2-aminobenzoic acid (12[C6]-2-
AA) and a 13[C6]-2-
aminobenzoic acid (13[C6]-2-AA) isotopic pair.
The reference standard used in the methods according to the invention
comprises glycans labeled with
a stable (second) isotope of the fluorescent label, preferably with a stably
heavy isotope of the
fluorescent label. Preferably, the reference standard comprises at least the
same glycan structures as
analyzed in the sample. The reference standard allows for analyzing the glycan
structures individually.
The recombinant glycoprotein analyzed in the methods of the invention is
preferably a fusion protein or
an antibody, more preferably an Fc-fusion protein or an antibody. In
embodiments where the
recombinant protein is an Fc-fusion protein or antibody the glycan containing
fragment of the
recombinant glycoprotein released from the solid support in step c) is
preferably an Fc domain.
In one embodiment the recombinant glycoprotein is an antibody and the variable
region of the antibody
binds to the affinity ligand immobilized on the solid support, preferably the
affinity ligand is an antigen.
In a preferred embodiment the antibody is an IgG antibody, more preferably an
IgG 1 or IgG2 antibody
and even more preferably a human or humanized IgG 1 or IgG2 antibody.
In another embodiment the recombinant glycoprotein is an Fc-fusion protein
comprising an Fc-domain
and an effector domain and the effector domain binds to the affinity ligand
immobilized on the solid
support, wherein the affinity ligand is a binding partner or an antibody
specifically binding to the effector
domain of the Fc-fusion protein.
The enzyme used in the methods of the invention to release the glycan
containing Fc-domain of the
recombinant glycoprotein from the solid support is preferably an
endoproteinase, such as papain, ficin,
cysteine protease SpeB (FabULOUS) or cysteine proteinase IdeS (FabRICATORO),
preferably the
endoproteinase is the cysteine proteinase IdeS (FabRICATORO).
In one embodiment of the methods of the invention, the recombinant protein is
an antibody or an Fc-
fusion protein and the pre-clearing step comprises aai) immobilizing the Fc-
containing glycoproteins on
a solid support using an Fc-binding protein, wherein said Fc-binding protein
is preferably selected from
6
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
protein G or protein A, more preferably said Fc-binding protein is protein G;
and aaii) eluting the Fc-
containing glycoproteins; wherein the solid support used in said pre-clearing
step is made of the same
material as the solid support used in step b), but is not coupled to said
affinity ligand.
The LC-MS used in the methods of the present invention is preferably a reverse
phase LC-MS or a
NanoLC-MS, more preferably the LC-MS is a reverse phase NanoLC-MS.
In certain embodiments of the methods of the present invention, the solid
support is a resin comprising
microbeads, preferably sepharose beads, agarose beads or magnetic beads, more
preferably
sepharose beads. The affinity ligand may be coupled to the solid support via N-
hydroxysuccinimide
(NHS), cyanogen bromide, epoxy, carbodiimide or thiopropyl, preferably via N-
hydroxysuccinimide
(NHS).
In a preferred embodiment of the methods of the present invention, the two or
more liquid samples of a
mammal are prepared for analysis using a multi well filter plate, preferably
in a 24, 96 or 384 well filter
plate, more preferably in a 96 well filter plate.
The liquid samples of a mammal comprising the recombinant protein is a body
fluid and is preferably
selected from the group consisting of serum or plasma, urine, cerebral spinal
fluid, amniotic fluid, saliva,
sweat, ejaculate, tears, phlegm, vaginal secretion, vaginal wash and colonic
wash; more preferably said
samples are plasma or serum samples. Preferably the samples to be analyzed
have a volume from
about 1 pl to about 1000 pl, from about 5 pl to about 500 pl, from about 10 pl
to about 200 pl, from
about 10 pl to about 100 pl, from about 25 pl to about 100 pl, or from about
40 pl to about 75 pl,
preferably of about 50 pl. In a preferred embodiment the two or more liquid
samples of a mammal were
obtained from the same subject. The mammalian liquid sample analyzed in the
methods of the
invention may be a human, a monkey, a rodent, a dog, a cat or a pig sample,
preferably a rodent
sample such as from mouse, rat, hamster or rabbit.
The methods of the present invention may be used to determine pharmacokinetic
parameters of at least
one specific glycan structure of the glycans of the recombinant glycoprotein
of interest, preferably to
determine the C
max, t -max, AUG or t,2 thereof.
In certain embodiment, the glycans are analyzed in an aliquot of each of said
two or more liquid
samples of a mammal and optionally the concentration of said recombinant
glycoprotein is analyzed in
a further aliquot of said two or more liquid samples of a mammal, e.g., by
ELISA.
Preferably the aliquots to be analyzed have a volume from about 1 pl to about
1000 pl, from about 5 pl
to about 500 pl, from about 10 pl to about 200 pl, from about 10 pl to about
100 pl, from about 25 pl to
about 100 pl, or from about 40 pl to about 75 pl, preferably of about 50 pl.
The present invention also relates to a method of preparing a glycoprotein
based pharmaceutical
composition comprising analyzing the glycans of a recombinant glycoprotein in
liquid samples of a
mammal according to the method of the invention and formulating the
glycoprotein into said
pharmaceutical composition.
7
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
In yet another aspect the invention relates to a method of analyzing glycans
of an Fc-fusion protein of
interest in liquid samples of a mammal, comprising a) providing two or more
liquid samples from a
mammal comprising an Fc-fusion protein containing an Fc-domain and an effector
domain; aa) pre-
clearing each of the two or more liquid samples of a mammal comprising:
immobilizing the Fc-fusion
protein of each of said samples on a separate solid support using an Fc-
binding protein, wherein said
Fc-binding protein is preferably selected from protein G or protein A, more
preferably said Fc-binding
protein is protein G; and eluting the Fc-fusion protein; b) immobilizing the
Fc-fusion protein of each of
said samples on a separate solid support coupled to an affinity ligand
specific for the effector domain of
the Fc-fusion protein in the sample, wherein the affinity ligand is a binding
partner or an antibody
specifically binding to the effector domain of the Fc-fusion protein; c)
releasing the Fc-domain of the Fc-
fusion protein of each of said samples from the solid support into separate
eluates by cleaving with an
endopeptidase specific for Fc-fusion proteins; dd) releasing the glycans from
the immobilized effector
domain of the Fc-fusion protein of each of said samples remaining on the solid
support following step c)
into separate solutions; ee) labeling the glycans of each of said samples with
a first stable isotope of a
fluorescent label; and ff) analyzing the labeled glycans of step ee) of each
of said samples separately
using LC-MS and comparing said two or more samples; wherein a reference
standard comprising
glycans labeled with a second stable isotope of the fluorescent label is added
to each of said samples
prior to step dd), ee) or ff) and wherein the reference standard is analyzed
together with the labeled
glycans of step ff).
Brief Description of the Figures
Figure 1: Schematic illustration of the work flow of the glycan PK profiling
method exemplified for Fc-
fusion proteins and antibodies. The work flow shows A) an Fc-fusion protein
specific sample
preparation and B) an antibody specific sample preparation, C) N-glycan
processing of the Fc-fusion
protein sample (1) and the antibody sample (2) and D) glycan clean-up after
labeling followed by RP
NanoLC-MS analysis.
Figure 2: A) Glycan map obtained from quality control (QC) samples showing the
relative N-glycan
composition (left) and a magnified view (right) showing the minor abundant N-
glycans. The error bars
represent the standard deviation. The two columns without error bars, G1F and
GOF, extend beyond
the graph in the magnified view. QC sample concentrations were between 10
pg/ml and 100 pg/ml. B)
Schematic overview on the analyzed glycan structures, wherein GIcNAc = filled
square, Fuc = triangle,
Man = dark circles and Gal = light circle.
Figure 3: Comparision of nanoLC-MS based mAb1 glycan PK data (filled diamonds)
and ELISA data
(filled squares). Independent results from four glycans relative to the
maximum of each curve are shown
to enable comparison (mean SD, n =15). In (A) PK profile of GOF (about 70%
of glycans), in (B) PK
profile of G1F, in (C) PK profile of G2F, in (D) PK profile of M6G0F/M5G1F (<
0.1% of glycans), in (E)
PK profiles of M3, in (F) PK profile of M3G1F and in (G) PK profile of M3GOF
are shown.
8
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Figure 4: Comparison of nanoLC-MS based mAb1 high mannose glycan PK profiles
obtained by
nanoLC-MS and ELISA profiles. (A) PK profile of M6 (filled diamonds) compared
to ELISA (filled
squares). (B) PK profile of M5 (filled diamonds) compared to ELISA (filled
squares).
Figure 5: Glycan Maps of mAb1 for each time point. Shown are mean percentage
of recovered mAb1
N-glycans after single subcutaneous administration in rabbits. Full view
(left) and magnified view (right)
are shown.
Figure 6: Fusion protein N-glycans of the two different batches (FP1 and FP2)
separated into Fc part
and receptor part. A) Glycan maps show the relative N-glycan composition of
FP1 Fc part (top left), FP1
receptor part (top right), FP2 Fc part (bottom left) and FP2 receptor part
(bottom right). The magnified
view shows the minor abundant N-glycans. B) Schematic overview on the analyzed
glycan structures,
wherein GIcNAc = filled square, Fuc = triangle, Man = dark circles, Gal =
light circle and SA = diamond.
Figure 7: PK profiles of FP1 and FP2 determined by ELISA. Error bars represent
the variability
between the animals (n=5).
Figure 8: Comparison of nanoLC-MS based fusion protein glycan PK profiles with
ELISA profiles.
Shown are the average PK profiles of the major N-glycans (A) of the Fc-domain
of FP1 and FP2 as
measured by NanoLC-MS (filled squares) and (B) of the receptor part (effector
domain) of FP1 and FP2
as measured by NanoLC-MS (filled squares) compared to the relative protein
concentration as
measured by ELISA (filled diamonds). Provided is the mean with the standard
deviation of 5 animals.
Figure 9: Glycan Maps for N-glycans located on the FP1 Fc part (left) and the
receptor part (right) are
shown separately.
Figure 10: Glycan Maps for N-glycans located on the FP2 Fc part (left) and the
receptor part (right) are
shown separately.
Figure 11: Influence of terminal sugar moieties. Percentages of terminal
sialylation (diamond), terminal
galactosylation (squares) and terminal N-acetylglucosamine (triangle) are
pictured (A) the for Fc-part of
fusion protein 1 (FP1) and fusion protein 2 (FP2) and (B) for the receptor
part (effector domain) of
fusion protein 1 (FP1) and fusion protein 2 (FP2).
Detailed Description of the Invention
As used herein, a "fluorescent label" is a fluorescent compound used in LC-MS
analysis. Typically
reductive amination is used to introduce a fluorescent label to
oligosaccharides, facilitating detection in
the subsequent separation, preferably such fluorescent label is 2-amino
benzamide (2-AB) or 2-amino
benzoic acid (2-AA). 2-AA is 2-amino benzoic acid, also known as anthranilic
acid. For the purposes of
the present invention the fluorescent label is available as an isotopic pair,
containing a first stable
isotope (e.g. light isotype) of a fluorescent label and a second stable
isotope (e.g., heavy isotype) of a
fluorescent label. Examples for such isotopic pairs useful in the present
invention is isotopic 2-AA, for
example, the stable isotopic variants 12[C6]-2-AA and 13[C6]-2-AA (Prien,
J.M., et al., Analytical
9
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Chemistry 82, 1498-508 (2010)). Glycans labeled with such isotopic variants of
for example 2-AA may
be analyzed using positive-mode matrix-assisted laser desorption ionization
time-of-flight mass
spectrometry (MALDI-TOF MS) or on-line negative-mode electrospray mass
spectrometry (ESI-
MS/MSn) and/or nanospray mass spectrometry (NSI-MS) (Prien, J.M., et al.,
Analytical Chemistry 82,
1498-508 (2010)). Other examples of a stable isotopic pair of a fluorescent
label are 12[C6]-aniline/
13[C6]-aniline and 12[C6]-2-AB/13[C6]-2-AB.
Carbohydrate moieties are described herein with reference to commonly used
nomenclature for
oligosaccharides. A review of carbohydrate chemistry which uses this
nomenclature can be found, for
example, in Hubbard and !vett, Ann. Rev. Biochem. 50, 555-583 (1981).
For the purposes of the present invention, a "glycan" refers to any sugar or
assembly of sugars (i.e.,
saccharide or carbohydrate), in free form or attached to another molecule,
such as proteins or lipids. As
referred to herein, an "N-glycan" is a glycan covalently linked to an
asparagine residue of a polypeptide
chain in the consensus sequence: -Asn-X-Ser/Thr (e.g., comprising the common
core structure Menai-
6(Mana1-3)Man81-4G1cNAc81-4GIcNAc81-N-Asn). As referred to herein, an "acidic
glycan" is an N-
glycan containing at least one terminal sialic acid (at the non-reducing
terminus). As referred to herein,
a "neutral glycan" is an N-glycan that does not contain any sialic acid. As
referred to herein,
"glycosylation" is the enzyme-catalyzed covalent attachment of a carbohydrate
to a polypeptide, lipid,
polynucleotide, carbohydrate, or other organic compound, generally catalyzed
by glycosyltransferases,
utilizing specific sugar nucleotide donor substrates. For the purposes of the
present invention, a
"polysaccharide" is a glycan composed of repeating monosaccharides, preferably
greater than ten
monosaccharide units in length. As referred to herein, a "sugar" refers to any
carbohydrate, preferably
to low molecular weight carbohydrates that are sweet in taste (see glossary of
Essentials of
Glycobiology. 2nd edition. Varki A, Cummings RD, Esko JD, et al., editors.
Cold Spring Harbor (NY):
Cold Spring Harbor Laboratory Press; 2009). The following abbreviations for
monosaccharides are
used: N-acetylglucosamine (GIcNAc), fucose (Fuc), mannose (Man), galactose
(Gal), and sialic acid
(SA).
For the purposes of the present invention, the term "glycoprotein" refers to
peptides and proteins,
including antibodies and fusion proteins (e.g., Fc-fusion proteins), having at
least one glycan side chain.
The term "recombinant glycoprotein" as used herein refers to a glycosylated
protein that results from
the expression of recombinant DNA within a living cell. Recombinant DNA
molecules are DNA
molecules formed by laboratory methods of genetic recombination, such as
molecular cloning, to bring
together genetic material from multiple sources, creating sequences that would
not otherwise be found
in biological organisms. In the present context, the recombinant glycoprotein
also includes that the
protein is not endogenous to the mammal of which the liquid sample was
obtained. In other words, the
recombinant glycoprotein has been administered to the mammal prior to
obtaining the sample.
Preferred recombinant glycoproteins are therapeutic recombinant glycoproteins
that have been
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
expressed in higher eukaryotic cells. In this regard, exemplary useful higher
eukaryotic cells are
selected from the following cell lines:
Cell Line Meaning Origin Tissue Origin
Morphology
CHO Chinese hamster ovary Hamster Ovary
Epithelium
COS-7 Cercopithecus aethiops, Ape - Kidney
Fibroblast
origin-defective SV-40 Cercopithecus
aethiops
(Chlorocebus)
BHK-21 Baby hamster kidney Hamster Kidney
Fibroblast
fibroblast cells
HEK-293 Human embryonic Human Kidney (embryonic)
Epithelium
kidney
HeLa "Henrietta Lacks" Human Cervical cancer
Epithelium
HL-60 Human leukemia Human Myeloblast Blood
cells
HUVEC Human umbilical vein Human Umbilical vein
Epithelial
endothelial cell endothelium
Jurkat Human T cell leukemia white
blood
cells
MCF-7 Michigan Cancer Human Mammary gland Invasive
breast
Foundation-7 ductal
carcinoma
NIH-3T3 NIH, 3-day transfer, Mouse Embryo
Fibroblast
inoculum 3 x 105 cells
RenCa Renal carcinoma Mouse Renal
carcinoma
U937 Human Leukaemic
monocytic lymphoma
Vero cells Vero (truth) African green Kidney epithelium
monkey
Particularly preferred in this regard are CHO cells or CHO derived cells
(e.g., CHO-DX611 or, CHO-
DG55, which are widely used in the art to express biopharmaceutically useful
glycoproteins, such as
antibodies.
The glycoproteins may be homologous to the host cell, or may be heterologous,
i.e., foreign, to the host
cell being utilized, such as, for example, a human glycoprotein produced by a
Chinese hamster ovary
(CHO) host cell. In certain embodiments, the glycoproteins expressed by a host-
cell are directly
secreted into the medium.
Examples of suitable mammalian, and in particular human, glycoproteins include
the following
molecules: a cytokine; a cytokine receptor; a chemokine, such as TNF and TECK;
a chemokine
receptor, such as a TNFR and CCR9; a growth hormone, such as human growth
hormone and bovine
11
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
growth hormone; growth hormone releasing factor; parathyroid hormone; thyroid
stimulating hormone; a
lipoprotein; alpha-1-antitrypsin; insulin A-chain; insulin B-chain;
proinsulin; follicle stimulating hormone;
calcitonin; luteinizing hormone; glucagon; human macrophage inflammatory
protein (MIP-1-alpha); a
serum albumin, such as human serum albumin; mullerian-inhibiting substance;
relaxin A-chain; relaxin
B-chain; prorelaxin; mouse gonadotropin-associated peptide; a clotting factor,
such as factor VIIIC,
factor IX, tissue factor, and von Willebrand factor; an anti-clotting factor,
such as Protein C; atrial
natriuretic factor; lung surfactant; a plasminogen activator, such as
urokinase or human urine or tissue-
type plasminogen activator (t-PA); bombesin; thrombin; hemopoietic growth
factor; enkephalinase;
RANTES; a microbial protein, such as beta-lactamase; DNase; inhibin; activin;
vascular endothelial
growth factor (VEGF); a receptor for hormones or growth factors; an integrin;
protein A or D; a
rheumatoid factor; a neurotrophic factor such as bone-derived neurotrophic
factor (BDNF),
neurotrophin-3, -4, -5, or -6 (NT-3, NT-4, NT-5, or NT-6); a nerve growth
factor, such as NGF-beta;
platelet-derived growth factor (PDGF); a fibroblast growth factor, such as
aFGF and bFGF; epidermal
growth factor (EGF); a transforming growth factor (TGF) such as TGF-alpha and
TGF-beta, including
TGF-beta1, TGF-beta2, TGF-beta3, TGF-beta4, or TGF-beta5; insulin-like growth
factor-land -II (IGF-I
and IGF-II); des(1-3)-IGF-1 (brain IGF-I); an insulin-like growth factor
binding protein, a CD protein, such
as CD-3, CD-4, CD-8, and CD-19; erythropoietin; an osteoinductive factor, an
immunotoxin; a bone
morphogenetic protein (BMP); an interferon, such as interferon-alpha, -beta,
and -gamma; a colony
stimulating factor (CSF), such as M-CSF, GM-CSF, and G-CSF; an interleukin
(ILs), such as IL-1 to IL-
21; superoxide dismutase; a T-cell receptor; a cell surface membrane protein;
a transport protein; a
homing receptor; a regulatory protein; a decay accelerating factor; and a
viral antigen, such as, a
portion of the HIV-1 or HIV-2 envelope protein.
A preferred glycoprotein in the context of the present invention is a fusion
protein. The term "fusion
protein" as used herein refers to a chimeric protein containing two proteins
or protein fragments fused to
each other (i.e., expressed as one polypeptide) often separated by an amino
acid linker, preferably an
effector domain fused to an Fc domain, albumin or transferring. Thus, the
fusion protein is preferably an
Fc-fusion protein, a transferrin-fusion protein or an albumin-fusion protein,
more preferably, the fusion
protein is an Fc-fusion protein. Typically the Fc-domain in a fusion protein
contains the CH2 and CH3
region and the hinge region of the IgG heavy chain, preferably of the IgG1
heavy chain. Fusion to an
Fc-domain, albumin or transferrin usually increases the in vivo half-life
and/or increases solubility of the
effector domain. The amino acid linker may further contain a cleavage site for
an endoproteinase. The
effector domain may be a full length therapeutically relevant protein or a
fragment thereof, particularly
the extracellular part of a therapeutically relevant receptor or a membrane
protein. Non limiting
examples of therapeutically relevant Fc-fusion proteins are LFA-3/Fc
(Alefecept), TNFR-Fc
(Etanercept), CTLA-4/Fc (Abatacept and Belatacept), VEGFR1/2-Fc (Aflibercept),
rFVII1Fc and rFIXFc,
IL-1R1-Fc (Rilonacept), thrombopoietin-Fc (Romiplostim) and angiopoietin-1/2
antagonist peptide-Fc
(Trebananib).
12
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Examples of suitable mammalian, and in particular human, effector proteins
suitable to be used in
fusion proteins include the following molecules: a cytokine; a cytokine
receptor; a chemokine, such as
TNF and TECK; a chemokine receptor, such as a TNFR and CCR9; a growth hormone,
such as human
growth hormone and bovine growth hormone; growth hormone releasing factor;
parathyroid hormone;
thyroid stimulating hormone; a lipoprotein; alpha-1-antitrypsin; insulin A-
chain; insulin B-chain;
proinsulin; follicle stimulating hormone; calcitonin; luteinizing hormone;
glucagon; human macrophage
inflammatory protein (MIP-1-alpha); mullerian-inhibiting substance; relaxin A-
chain; relaxin B-chain;
prorelaxin; mouse gonadotropin-associated peptide; a clotting factor, such as
factor VIIIC, factor IX,
tissue factor, and von Willebrand factor; an anti-clotting factor, such as
Protein C; atrial natriuretic
factor; lung surfactant; a plasminogen activator, such as urokinase or human
urine or tissue-type
plasminogen activator (t-PA); bombesin; thrombin; hemopoietic growth factor;
enkephalinase; RANTES;
a microbial protein, such as beta-lactamase; DNase; inhibin; activin; vascular
endothelial growth factor
(VEGF); a receptor for hormones or growth factors; an integrin; protein A or
D; a rheumatoid factor; a
neurotrophic factor such as bone-derived neurotrophic factor (BDNF),
neurotrophin-3, -4, -5, or -6 (NT-
3, NT-4, NT-5, or NT-6); a nerve growth factor, such as NGF-beta; platelet-
derived growth factor
(PDGF); a fibroblast growth factor, such as aFGF and bFGF; epidermal growth
factor (EGF); a
transforming growth factor (TGF) such as TGF-alpha and TGF-beta, including TGF-
beta1, TGF-beta2,
TGF-beta3, TGF-beta4, or TGF-beta5; insulin-like growth factor-I and -II (IGF-
I and IGF-II); des(I-3)-
IGF-I (brain IGF-I); an insulin-like growth factor binding protein, a CD
protein, such as CD-3, CD-4, CD-
8, and CD-19; erythropoietin; an osteoinductive factor, an immunotoxin; a bone
morphogenetic protein
(BMP); an interferon, such as interferon-alpha, -beta, and -gamma; a colony
stimulating factor (CSF),
such as M-CSF, GM-CSF, and G-CSF; an interleukin (ILs), such as IL-1 to IL-21;
superoxide
dismutase; a T-cell receptor; a cell surface membrane protein; a transport
protein; a homing receptor; a
regulatory protein; a decay accelerating factor; and a viral antigen, such as,
a portion of the HIV-1 or
HIV-2 envelope protein.
Another preferred glycoprotein in the context of the present invention is an
antibody. The term
"antibody" as referred to herein includes whole antibodies and any antigen
binding fragment (i.e.,
antigen-binding portion) or single chain thereof. An "antibody" refers to a
glycoprotein comprising at
least two heavy (H) chains and two light (L) chains inter-connected by
disulfide bonds, or an antigen
binding portion thereof. Each heavy chain is comprised of a heavy chain
variable region (abbreviated
herein as VH) and a heavy chain constant region. The heavy chain constant
region is comprised of
three domains, CH1, CH2 and CH3. Each light chain is comprised of a light
chain variable region
(abbreviated herein as VL) and a light chain constant region. The light chain
constant region is
comprised of one domain, CL. The VH and VL regions can be further subdivided
into regions of
hypervariability, termed complementarity determining regions (CDR),
interspersed with regions that are
more conserved, termed framework regions (FR). Each VH and VL is composed of
three CDRs and four
FRs, arranged from amino-terminus to carboxy-terminus in the following order:
FR1, CDR1, FR2,
CDR2, FR3, CDR3, FR4. The variable regions of the heavy and light chains
contain a binding domain
13
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
that interacts with an antigen. The constant regions of the antibodies may
mediate the binding of the
immunoglobulin to host tissues or factors, including various cells of the
immune system (e.g., effector
cells) and the first component (C1q) of the classical complement system.
Typically antibodies are
glycosylated at the CH2 domain at asparagine residue N297. A significant
number of antibodies also
possess additional glycosylation sites (i.e., the Asn-X-Ser/Thr tripeptide) in
their variable regions and N-
linked glycosylation can be found in variable (V) domains of both heavy (VH)
and light (VL) chains of
serum IgG and of some monoclonal antibodies (mAbs). Examples of antigen-
binding fragments
encompassed within the term "antigen-binding portion" of an antibody include
(i) a Fab fragment, a
monovalent fragment consisting of the VL, VH, CL and CH1 domains; (ii) a
F(ab')2 fragment, a bivalent
fragment comprising two Fab fragments linked by a disulfide bridge at the
hinge region; (iii) an Fd
fragment consisting of the VH and CH1 domains; (iv) an Fv fragment consisting
of the VL and VH
domains of a single arm of an antibody, (v) a dAb fragment (Ward et al.,
Nature. 341:544-546 (1989)),
which consists of a VH domain; and (vi) an isolated complementarity
determining region (CDR) or (vii) a
combination of two or more isolated CDRs which may optionally be joined by a
synthetic linker.
Furthermore, although the two domains of the Fv fragment, VL and VH, are coded
for by separate
genes, they can be joined, using recombinant methods, by a synthetic linker
that enables them to be
made as a single protein chain in which the VL and VH regions pair to form
monovalent molecules
(known as single chain Fv (scFv); see e.g., Bird et al. Science 1988, 242:423-
426; and Huston et al.,
Proc. Natl. Acad. Sci. USA 1988, 85:5879-5883). Such single chain antibodies
are also intended to be
encompassed within the terms antigen-binding portion and antigen-binding
fragment of an antibody.
These antibody fragments are obtained using conventional techniques known to
those with skill in the
art, and the fragments are screened for utility in the same manner as are
intact antibodies.
The term "monoclonal antibody" (mAb), as used herein, refers to an antibody
which displays a single
binding specificity and affinity for a particular epitope. Accordingly, the
term "human monoclonal
antibody" refers to an antibody which displays a single binding specificity
and which has variable and
constant regions derived from human germ line immunoglobulin sequences. In one
embodiment,
human monoclonal antibodies are produced by a hybridoma which includes a B
cell obtained from a
transgenic non-human animal, e.g., a transgenic mouse, having a genome
comprising a human heavy
chain transgene and a light chain transgene fused to an immortalized cell.
The term "recombinant human antibody", as used herein, includes all human
antibodies that are
prepared, expressed, created or isolated by recombinant means, such as (a)
antibodies isolated from
an animal (e.g., a mouse) that is transgenic or transchromosomal for human
immunoglobulin genes or a
hybridoma prepared therefrom, (b) antibodies isolated from a host cell
transformed to express the
antibody, e.g., from a transfectoma, (c) antibodies isolated from a
recombinant, combinatorial human
antibody library, and (d) antibodies prepared, expressed, created or isolated
by any other means that
involve splicing of human immunoglobulin gene sequences to other DNA
sequences. Such recombinant
human antibodies have variable and constant regions derived from human
germline immunoglobulin
sequences. In certain embodiments, however, such recombinant human antibodies
can be subjected to
14
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
in vitro mutagenesis (or, when an animal transgenic for human Ig sequences is
used, in vivo somatic
mutagenesis) and thus the amino acid sequences of the VH and VL regions of the
recombinant
antibodies are sequences that, while derived from and related to human
germline VH and VL sequences,
may not naturally exist within the human antibody germline repertoire in vivo.
As used herein, a "heterologous antibody" is defined in relation to the
transgenic non-human organism
producing such an antibody. This term refers to an antibody having an amino
acid sequence or an
encoding nucleic acid sequence corresponding to that found in an organism not
consisting of the
transgenic non-human animal, and generally from a species other than that of
the transgenic non-
human animal.
The term "peptide" as used herein refers to chains of amino acid monomers
linked by peptide amide
bonds. Peptides may be short chains of about 10 amino acids or less. Peptides
may also be long
chains of about 70 amino acids or more or anything in between. In the context
of the present invention
the fragments of a recombinant glycoprotein released by an endoproteinase is
referred to as a peptide
or glycopeptide if it carries a glycan.
As used herein "affinity ligand" refers to a ligand that binds specifically to
the recombinant glycoprotein
of interest. This affinity ligand is used to separate the recombinant
glycoprotein of interest from the
other compounds in the mammalian liquid sample, such as serum or plasma, using
affinity
chromatography. Typically, if the recombinant glycoprotein is an antibody, the
affinity ligand is an
antigen. For other recombinant proteins, the affinity ligand is typically a
binding partner or a substrate of
said recombinant protein. If the recombinant protein is a fusion protein the
affinity ligand is typically a
binding partner or a substrate of the effector domain of the fusion protein,
e.gõ the ligand for a receptor
or vice versa. Alternatively the affinity ligand may be an antibody specific
for said recombinant protein or
an antibody specific for said effector domain of the fusion protein.
Preferably, the affinity ligand is
aglycosylated, i.e., carries no glycans or at least no N-glycans, particularly
if the glycans of the
immobilized glycan containing fragment of the recombinant glycoprotein is
further analyzed. The skilled
person will understand that Protein A and Protein G or any other generically
Fc-domain binding protein
is not suitable as an affinity ligand for a specific recombinant Fc-fusion
protein or antibody of interest in
a liquid samples of a mammal containing endogenous antibodies, particularly
for serum and plasma.
The term "antigen" as uses herein is a substance which provokes an adaptive
immune response and
may also be referred to as immunogen. An antigen binds to the antigen-binding
site of an antibody.
Antigens are typically of high molecular weight and proteins or
polysaccharides. As used herein an
antigen may also refer to the immunogenic part of an antigen comprising the
epitope, e.g., peptides.
Peptides, lipids, nucleic acids and many other materials can also function as
antigens. Immune
responses may also be generated against smaller substances, called haptens, if
these are chemically
coupled to a larger carrier protein, such as bovine serum albumin, keyhole
limpet hemocyanin (KLH) or
other synthetic matrices.
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
As used herein, "specific binding", "selective binding" and "selectively
binds" and "ligand specific for",
refer to an affinity ligand, binding to a predetermined recombinant
glycoprotein, such as the effector
domain of a fusion protein or to an antibody. For example, in one embodiment,
the ligand binds with an
affinity (KD) of approximately less than 10-7 M, such as approximately less
than 10-8 M, 10-9 M or 10-
10 M or even lower to the recombinant glycoprotein when determined by surface
plasmon resonance
(SPR) technology in a BIACORE 3000 instrument, and binds to the predetermined
recombinant
glycoprotein with an affinity that is at least two-fold greater than its
affinity for binding to a non-specific
antigen or other glycoproteins in the liquid sample. Thus, in the context of
the invention "a ligand
specific for" means that the ligand preferentially binds to the recombinant
glycoprotein and thus
separates said recombinant glycoprotein from other glycoproteins present in
the sample, particularly
from other antibodies present in the sample. The phrases "an antibody
recognizing an antigen" and "an
antibody specific for an antigen" are used interchangeably herein with the
term "an antibody which
selectively binds to an antigen" and may equally be used vice versa that "an
antigen is specific for an
antibody".
The term "eluate" as used herein relates to the material released from the
solid support. It may
comprise the buffer or solution used during elution or a different buffer or
solution.
As used herein, "LC" as used herein means liquid chromatography, which may be
combined with mass
spectrometry, referred to as "LC-MS" herein. LC-MS may further include tandem
MS. The term
"reversed phase LC" or "HPLC" (RP-LC or RP-HPLC) as used herein has a non-
polar stationary phase
and an aqueous, moderately polar mobile phase, e.g., a silica which has been
surface-modified with
RMe2SiCI, where R is a straight chain alkyl group such as C181-137 or C81-117
(Lehto and Hou, Chemistry
and Analysis of Radionuclides, Wiley-VCH Verlag & Co., Weinheim, Germany,
2011, page 170). In the
context of the present invention "reverse phase LC-MS" is preferred as the
mobile phase used in
reverse phase is more compatible with MS and therefore more sensitive than
normal phase LC-MS.
As used herein "nano-LC" or nano-HPLC (RP-nano-LC or RP-nano-HPLC) is
characterized by a
decreased inner diameter of the columns that are used for LC (10-150 pm) and
smaller flow-rates (10-
1000 nl/min) compared to conventional LC or HPLC, respectively. This down-
scaling results in high
plate counts of the nano-LC system and the ability to analyze proteinaceous
samples in the low
femtomole and subfemtomole ranges (Chervet et al., Analytical Chemistry 1996,
68:1507-12).
Consistent with the understanding and common general knowledge in the field of
liquid
chromatography, it will be appreciated that in accordance with the invention
nano-LC and nano-HPLC
are suitable and intended forms of LC and HPLC, respectively, for the purposes
of the present
invention; and that RP-nano-LC and RP-nano-HPLC are suitable and intended
forms and even
preferred forms of RP-LC and RP-HPLC, respectively. The same applies to the
likewise well-known and
established techniques of micro-LC and capillary-LC, or RP-micro-LC and RP-
capillary-LC, which for
the purposes of the present invention are suitable and intended forms of LC,
and RP-LC, respectively.
16
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Where the terms LC, HPLC, LC-MS or HPLC-MS is used herein, this also
encompasses their preferred
embodiments, nano-LC, nano-HPLC, nano-LC-MS or nano-HPLC-MS and their reverse
phase forms.
For the purposes of the present invention, a "mobile phase" of RP-LC or RP-
HPLC is preferably a
gradient of an organic modifier (e.g., acetonitrile or methanol) in water,
with an ionic modifier that
controls the pH and ionization state or acts as an ion pairing reagent.
Anionic ion-pair reagents (e.g.,
trifluoroacetic acid (TFA)) bind to protonated basic groups of peptides. The
addition of 0.1% TFA
acidifies the eluent which causes the carboxylic groups of peptides and
proteins to become protonated,
resulting in a larger hydrophobicity of the molecules. Cationic ion-pairing
reagents (e.g.,
triethylammonium ions) bind to ionized carboxyl groups of peptides ("Protein
Liquid Chromatography',
Journal of Chromatography Library, vol. 61, edited by Kastner M, Elsevier
Science By., 2000, page
153). Diethylamine (DEA) can also be used as an ion pairing reagent (Melmer et
al., Journal of
Chromatography A (2011), Volume: 1218(1): 118-123). For normal phase or HILIC
chromatography of
e.g. 2-AB labeled N-glycans, a suitable mobile phase consists for example of
60 mMol ammonium
formate in 75% acetonitrile (mobile phase A) and 115 mMol ammonium formate in
54% acetonitrile
(mobile phase B) (Melmer et al., Anal Bioanal Chem (2010), Volume 398: 905-
914).
As used herein, "ion trap mass spectrometry" is an arrangement in which ions
with a desired range of
quotients mass/charge are first made to describe stable paths under the effect
of a high-frequency
electric quadrupole field, and are then separated and presented to a detector
by adjusting the field so
as to selectively induce path instability according to their respective
mass/charge ratios e.g., quadrupole
ion trap (see http://www.genomicglossaries.com/content/mass spectrometry.asp).
Sensitivity of the
methods of the invention can be improved by using more sensitive nano ESI
sources.
For the purposes of the present invention, all "N-glycans" are understood to
have the common core
sugar sequence, Mana1-6(Mana1-3)Man61-4G1cNAc61-4GIcNAc61-Asn-X-Ser/Thr, and
are
classified into three types: (1) "oligomannose" (high mannose), in which only
mannose residues are
attached to the core; (2) "complex", in which "antennae" initiated by N-
acetylglucosaminyltransferases
(GIcNAcTs) are attached to the core; and (3) "hybrid", in which only mannose
residues are attached to
the Mana1-6 arm of the core and one or two antennae initiated by N-
acetylglucosaminyltransferases
(GIcNAcTs) are on the Mana1-3 arm (Essentials of Glycobiology, 2nd edition,
Varki et al., editors, Cold
Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 2009; Chapter 8).
For the purposes of the present invention, a reference to the analysis of
acidic or neutral glycans, N-
glycans, or glycans in general, "in one run", "in one approach" or "in a
single (analytical) approach"
means that the MS analysis is directly coupled to (or in other words, is
performed on-line with) the LC or
RP-LC step. This means that there is no further analytical or preparatory step
between the RP-LC and
the MS analysis. It will be understood that this does not exclude that
fluorescence detection will occur
between the RP-LC and the MS. Indeed, fluorescence detection of the 2-AA
labeled glycans separated
on the RP-column will normally (and advantageously) occur before they leave
the LC system and
17
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
proceed to MS, since the fluorescence detector (FLD) that is passed by the
labeled glycans is typically
a module of the HPLC/LC system. This is also referred to as "LC-MS" or RP-LC-
MS".
The term "enzymatic cleavage" as used herein refers to any cleavage of a
glycoprotein that involves an
enzyme and allows for release of the recombinant glycoprotein while avoiding
elution of most non-
specifically bound glycoproteins. The enzymatic cleavage may be done using an
endoglycanase or an
endoproteinase, wherein endoproteinases are preferred. The enzyme should be
selective for the
recombinant glycoprotein of interest. This means that the specific cleavage
site is rare and preferably
only present in the recombinant glycoprotein of interest or additionally in a
few other glycoproteins, such
as a specific family of glycoproteins, or a group of glycoproteins all
comprising the same domain. A
more selective enzyme avoids elution of fragments of non-specifically bound
glycoproteins from the
solid support. Preferably the enzyme cleaves the recombinant glycoprotein
efficiently and specifically
(e.g., only at a known cleavage site). Suitable enzymes are for example
endoproteinases that cleave
within the recombinant glycoprotein of interest specifically at a certain
consensus sequence and that
have narrow substrate specificity (or high selectivity). Preferably, the
endoproteinase is specific (and/or
selective) for the recombinant glycoprotein of interest or a domain contained
in the recombinant
glycoprotein of interest. For example, a suitable endoproteinase specifically
only cleaves the
recombinant glycoprotein of interest or a group of proteins comprising
essentially the same domain,
including the recombinant glycoprotein of interest. Examples for a group of
proteins comprising
essentially the same domain are antibodies and Fc-fusion proteins, all
containing an Fc-domain. A
suitable enzyme has preferentially only one cleavage site in the recombinant
glycoprotein of interest or
within each identical polypeptide chain within the recombinant glycoprotein of
interest. The terms
"endopeptidase", endoproteinase", "proteinase" or "protease" are used
interchangeably herein.
Preferably, the released glycan containing fragment is a peptide containing
one glycan or in the case of
a released Fc-domain two identical peptides containing one glycan each. For
example endoproteinases
suitable in the methods of the invention to release the glycan containing Fc-
domain of the recombinant
glycoprotein from the solid support are endoproteinase, such as papain, ficin,
cysteine protease SpeB
(FabULOUS) or cysteine proteinase IdeS (FabRICATORO), preferably the
endoproteinase is the
cysteine proteinase IdeS (FabRICATORO). For example IdeS specifically cleaves
human IgG in the
hinge region between the two glycines of the constant sequence ELLGGPS and
SpeB cleaves in the
hinge region between threonine and cysteine within the sequence KTHTCPPC. The
enzymatic
cleavage may also be done using a glycanase with specificity for certain N- or
0-glycosylation sites.
Such a glycanase is for example EndoSGlycanase (IgGZEROO), which is specific
for N-glycans in IgG
Fc-domain containing glycoproteins, which are cleaved after the first GIcNAc
at the reducing end.
Alternatively, fusion proteins such as Factor IX-albumin (FIX-albumin)
comprise a linker sequence
between the two domains that is based on amino acids 137-153 derived from the
N-terminus of the
activation peptide of FIX. Activation of FIX-recombinant albumin fusion
protein by either FXIa or
FVIIa/TF cleaves the linker, thereby separating the FIXa and rHA moieties of
the fusion protein. Thus,
18
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
FXIa and/or FVIIa/TF are suitable to cleave FIX-albumin selectively. Also
Factor Xa is a site-specific
protease that exhibit very low non-specific cleavage under many conditions.
Thus, Factor Xa may be
used for cleaving glycoproteins that contain a Factor Xa cleavage site
(cleavage behind arginine of the
sequence Ile-Glu/Asp-Gly-Arg). Factor Xa kits are e.g. available from Novagen
(69036-3). Likewise
thrombin is a site-specific protease for specific cleavage between arginine
and glycine of the sequence
LeuValProArgGlySer in recombinant fusion proteins.
The term "solid support" as used herein refers to any solid surface that can
be used to immobilize a
ligand thereon, particularly in affinity chromatography. Typically a "solid
support" suitable for affinity
chromatography is a resin, such as a resin comprising microbeads with a large
surface area.
Particularly suitable resins in the present invention are sepharose beads,
agarose beads or magnetic
beads, wherein sepharose beads are preferred. The term "immobilized" refers to
covalent or non-
covalent binding to the solid support, either directly or indirectly.
Typically the affinity ligand is covalently
coupled to the solid support, for example via N-hydroxysuccinimide (NHS),
cyanogen bromide, epoxide,
carboiimide or thiopropyl reactive groups. Commercially available microbeads
activated with one of the
above reactive groups are known in the art, e.g., NHS-activated sepharose.
Once the recombinant
glycoprotein binds to its specific affinity ligand immobilized or coupled to
the solid support the
recombinant glycoprotein is immobilized on the solid support via non-covalent
binding to the affinity
ligand.
The term "liquid sample of a mammal" as used herein refers to any liquid
sample obtained from a
mammal at a certain time point comprising biological material such as cells,
protein, DNA or RNA.
Typically the liquid sample is a body fluid, such as serum or plasma. However,
the liquid sample may
also be whole blood, urine, cerebral spinal fluid, amniotic fluid, saliva,
sweat, ejaculate, tears, phlegm,
vaginal secretion, vaginal wash or colonic wash. Preferably the sample is
cleared from cells or debris,
e.g., by centrifugation prior to be subject to the methods of the invention.
For the liquid sample to
contain the recombinant glycoprotein of interest, the sample must have been
obtained following
administration of said recombinant glycoprotein to the mammal. In the context
of the present invention
the recombinant glycoprotein is to be understood as a biopharmaceutical or
therapeutically active
recombinant glycoprotein. The skilled person will understand that a liquid
sample, or an "aliquot" of a
liquid sample, may be used in the present invention. If only an aliquot of the
liquid sample is used to be
analyzed according to the methods of the present invention, other aliquots of
the same sample may be
analyzed for different parameters, such as protein concentration, or stored
frozen for later analysis.
The term "lower limit of quantification (LLOQ)" as used herein refers to the
lowest concentration of an
analyte in a sample that can be quantitatively determined with suitable
precision and accuracy", wherein
the analyte is the recombinant glycoprotein. The LLOQ is typically given as
pg/ml. the skilled person
understands that it is therefore strongly dependent on the sample volume used
for analysis.
The term "reference standard" as used herein refers to a standardized mixture
of glycans with a known
relative distribution labeled with an isotopic variant of a fluorescent label
different to the isotopic variant
19
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
used for labeling the glycans to be analyzed. Isotope labeling can be
introduced into glycans at the
reducing end, for example using a stable (preferably heavy) isotopic variant
of the fluorescent label, e.g.
13[C6]-2AA. The mixture of glycans contains at least the glycan structures to
be analyzed. The mixture is
further standardized, which means that the identical mixture of glycans is
added to each sample of the
experiment or to each sample that is compared to each other. Preferably, the
glycans of one batch of
the recombinant glycoprotein are released, labeled and stored frozen until
use.. The use of a reference
standard compensates variations in the sample preparation and nanoLC-MS
analysis resulting in more
precise results. It is therefore preferably added as early as possible to the
sample. Moreover, the
reference standard allows for analyzing the glycan structures individually.
This means the amount of
each glycan structure can be quantified individually relative to the
respective glycan structure of the
reference standard based on the known relative distribution of the individual
glycans in the reference
standard, rather than the relative proportion of the glycans within each
sample. Hence, PK parameter of
individual glycan structures can be determined.
The term "high throughput" as used herein relates to a mode or method that
permits rapid and highly
parallel sample preparation of a number of samples, such as more then 10, more
than 50, more than
100, more than 500 or more than 1000 samples.
For the purposes of the present invention, a reference to % will be understood
to refer to (v/v) unless
explicitly stated otherwise.
As used herein, the term "about" when used together with a numerical value
(e.g., a pH value or a
percentage value) is intended to encompass a deviation of 20%, preferably 10%,
more preferably 5%,
even more preferably of 2%, and most preferably of 1% from that value. When
used together with a
numerical value it is at the same time to be understood as individually
disclosing that exact numerical
value as a preferred embodiment in accordance with the present invention.
As used herein, the term "comprising" is to be construed as encompassing both
"including" and
"consisting of", both meanings being specifically intended, and hence
individually disclosed,
embodiments in accordance with the present invention.
Accordingly, the present invention relates to a method of analyzing glycans of
a recombinant
glycoprotein of interest in liquid samples of a mammal comprising a) providing
two or more liquid
samples of a mammal comprising the recombinant glycoprotein of interest; b)
immobilizing the
recombinant glycoprotein of each of said samples on a separate solid support
coupled to an affinity
ligand specific for the recombinant glycoprotein in the samples; c) releasing
a glycan containing
fragment of the recombinant glycoprotein of each of said samples from the
solid support into separate
eluates by enzymatic cleavage of the recombinant glycoprotein; d) optionally
releasing the glycans from
the glycan containing fragment of the recombinant glycoprotein of each of said
samples in the separate
eluates; e) labeling the glycans of each of said samples with a first stable
isotope of a fluorescent label;
and f) analyzing the labeled glycans of step e) of each of said samples
separately using LC-MS and
comparing said two or more samples, wherein a reference standard comprising
glycans labeled with a
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
second stable isotope of the fluorescent label is added to each of said
samples prior to step c), d), e) or
f) and the reference standard is analyzed together with the labeled glycans of
step f). It is preferred that
the reference standard is added early during the process. Thus, the reference
standard may be added
prior to step f), preferably prior to step e) or prior to step d) and even
more preferably prior to step c). All
steps of the method of the invention are performed separately for each of the
two or more liquid
samples obtained from a mammal.
The present invention also relates to a method further comprising analyzing
the glycans of a second
glycan containing fragment of the recombinant glycoprotein of each of said
samples that remains
immobilized to the solid support in step c), comprising the following
additional steps: aa) pre-clearing
each of the two or more liquid samples of a mammal to remove glycoproteins
binding non-specifically to
the solid support prior to step b); dd) releasing the glycans from the
immobilized second glycan
containing fragment of the recombinant glycoprotein of each of said samples
remaining on the solid
support following step c) into separate solutions; ee) labeling the glycans of
each of said samples with a
first stable isotope of a fluorescent label; and if) analyzing the labeled
glycans of step ee) of each of
said samples separately using LC-MS and comparing said two or more samples,
wherein a reference
standard comprising glycans labeled with a second stable isotope of the
fluorescent label is added to
each of said samples prior to step dd), ee) or if) and wherein the reference
standard is analyzed
together with the labeled glycans of step if). The pre-clearing step aa) may
comprise contacting the
samples with a solid support, wherein the solid support is made of the same
material as the solid
support used in step b), but is not coupled to said affinity ligand. It is
preferred that the reference
standard is added early during the process. Thus, the reference standard may
be added prior to
analyzing the labeled glycans using LC-MS (step if), preferably prior to the
labeling of the glycans to be
analyzed (step ee) and more preferably added on-column prior to release of the
glycans of the effector
domain (step dd). In cases where the glycans of the second glycan containing
fragment of the
recombinant glycoprotein of each of said samples that remain immobilized to
the solid support in step c)
are further analyzed, preferably no reference standard has been added to each
of said samples prior to
step c). The skilled person will understand that all steps of the method of
the invention are performed
separately for each of the two or more liquid samples obtained from a mammal.
Preferably, the fluorescent labeling is done after the glycans have been
released from the glycoprotein
fragment. Release of the glycans may be done chemically or enzymatically by
methods well known to
those skilled in the art.
Affinity purification from liquid samples of a mammal, such as a serum
samples, may require a pre-
clearing step. A per-clearing step is particularly advantageous, if the
glycans of the immobilized second
glycan containing fragment following enzymatic cleavage are to be analyzed.
The pre-clearing step may
comprise contacting the samples with a solid support, wherein the solid
support is made of the same
material as the solid support used for immobilizing of the recombinant
glycoprotein (step b) without the
coupled affinity ligand. Non-specifically binding glycoproteins or glycolipids
remain bound to the solid
21
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
support. Thus, the recombinant glycoprotein may be in the flow through, i.e.,
the material not binding to
the solid support. Alternatively, the pre-clearing step may be an upstream
second affinity purification
step immobilizing the recombinant glycoprotein on the solid support followed
by eluting the recombinant
glycoprotein from the solid support.
In another aspect the invention relates to a method of analyzing glycans of an
Fc-fusion protein of
interest in liquid samples of a mammal, comprising a) providing two or more
liquid samples from a
mammal comprising an Fc-fusion protein containing an Fc-domain and an effector
domain; aa) pre-
clearing each of the two or more liquid samples of a mammal comprising:
immobilizing the Fc-fusion
protein of each of said samples on a separate solid support using an Fc-
binding protein, wherein said
Fc-binding protein is preferably selected from protein G or protein A, more
preferably said Fc-binding
protein is protein G; and eluting the Fc-fusion protein; b) immobilizing the
Fc-fusion protein of each of
said samples on a separate solid support coupled to an affinity ligand
specific for the effector domain of
the Fc-fusion protein in the sample, wherein the affinity ligand is a binding
partner or an antibody
specifically binding to the effector domain of the Fc-fusion protein; c)
releasing the Fc-domain of the Fc-
fusion protein of each of said samples from the solid support into separate
eluates by cleaving with an
endopeptidase specific for Fc-fusion proteins; dd) releasing the glycans from
the immobilized effector
domain of the Fc-fusion protein of each of said samples remaining on the solid
support following step c)
into separate solutions; ee) labeling the glycans of each of said samples with
a first stable isotope of a
fluorescent label; and ff) analyzing the labeled glycans of step ee) of each
of said samples separately
using LC-MS and comparing said two or more samples, wherein a reference
standard comprising
glycans labeled with a second stable isotope of the fluorescent label is added
to each of said samples
prior to step dd), ee) or ff) and wherein the reference standard is analyzed
together with the labeled
glycans of step ff).
According to the methods of the present invention the glycans to be analyzed
are preferably N-glycans.
N-glycans to be analyzed may be high mannose type, hybrid type or complex type
N-glycans.
Alternatively the glycans to be analyzed may also be 0-glycans.
The fluorescent label as used in the methods of the invention may be any label
suitable for LC-MS
analysis such as 2-amino benzamide (2-AB), 2-amino benzoic acid (2-AA) or 8-
aminonachthalene-1, 3,
6-trisulfonic acid (ANTS). However, there must be a stable light and a stable
heavy isotope variant
available to label the glycans to be analyzed and the reference standard with
a different isotope variant
of the same fluorescent label. Preferably, the fluorescent label is a 12[C]
and 13[C] isotopic pair, more
preferably a 12[C6] and 13[C6] isotopic pair, and even more preferably a
12[C6]-2-aminobenzoic acid
.12
k [C6]-2-AA) and a 13[C6]-2-aminobenzoic acid (13[C6]-2-AA) isotopic pair.
It will be appreciated that the labeling with 2-AA in accordance with the
methods of the present
invention is particularly advantageous for mapping (i.e., analyzing) neutral
or acidic glycans of a
recombinant glycoprotein, such as an antibody or a fusion protein. The glycans
to be analyzed are
preferably N-glycans, particularly N-glycans that are either of the
oligomannose, hybrid or complex type.
22
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
In one embodiment the N-glycans are G1F isomers, particularly the G1F isomers
with 1,3 or 1,6
galactosylation.
The reference standard used in the methods according to the invention
comprises glycans labeled with
a stable (second) isotope of the fluorescent label, preferably with a stably
heavy isotope of the
fluorescent label. Preferably, the reference standard comprises at least the
same glycan structures as
analyzed in the sample. If the glycans to the analyzed are in the form of
glycan containing fragments of
the recombinant glycoprotein, such as glycopeptides, the reference standard
equally comprises glycans
in the form of the glycan containing fragment of the recombinant glycoprotein,
such as glycopeptides.
The use of a reference standard compensates variations in the sample
preparation and nanoLC-MS
analysis, resulting in more precise results. It is therefore preferably added
as early as possible to the
sample. Moreover, the reference standard allows for analyzing the glycan
structures individually as the
samples can be normalized. This means the amount of each glycan structure can
be quantified
individually relative to the respective glycan structure of the reference
standard based on the known
relative distribution of the individual glycans in the reference standard.
Hence, PK parameter of
individual glycan structures can be determined. The reference standard may be
prepared by releasing
the glycans of the recombinant glycoprotein of interest and labeling the
glycans with one of the stable
isotope variants of the fluorescent label. The reference standard may be the
same or different in steps
e) and ee) in the present invention. Preferably the reference standard in
steps e) and ee) is the same.
For example for analyzing the glycans of a specific fusion protein both the
reference standard in steps
e) and ee) may comprise glycans released from said full-length fusion protein.
Alternatively the
reference standard in step e) may comprise the glycans released from the
fusion partner, such as the
Fc-domain, albumin or transferrin, of the fusion protein and the reference
standard in step ee) may
comprise the glycans released from the effector domain, such as the
extracellular domain of TNFR, of
the fusion protein. However, the reference standard must contain the glycan
structure of the glycans to
be analyzed and the relative proportion of the glycan structures within the
glycans of the reference
standard should be known. Using a reference standard as described above allows
for determining the
relative amount of each glycan structure separately.
The recombinant glycoprotein analyzed in the methods of the invention is
preferably a fusion protein or
an antibody, more preferably an Fc-fusion protein or an antibody. In
embodiments where the
recombinant protein is an Fc-fusion protein or antibody the glycan containing
fragment of the
recombinant glycoprotein released from the solid support in step c) is
preferably an Fc domain.
For Fc-domain containing glycoproteins, the method of the invention may
comprises the following steps:
a) providing two or more liquid samples of a mammal comprising the recombinant
glycoprotein of
interest having an Fc-domain; b) immobilizing the recombinant glycoprotein of
each of said samples on
a separate solid support coupled to an affinity ligand specific for the
recombinant glycoprotein in the
samples; c) releasing a glycan containing fragment of the recombinant
glycoprotein of each of said
samples, preferably the Fc-domain, from the solid support into separate
eluates by enzymatic cleavage
23
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
of the recombinant glycoprotein; d) optionally releasing the glycans from the
Fc-domain of the
recombinant glycoprotein in the separate eluates; e) labeling the glycans of
each of said samples with a
first stable isotope of a fluorescent label; and f) analyzing the labeled
glycans of step e) of each of said
samples separately using LC-MS and comparing said two or more samples, wherein
a reference
standard comprising glycans labeled with a second stable isotope of the
fluorescent label is added to
each of said samples prior to step c), d), e) or f) and the reference standard
is analyzed together with
the labeled glycans of step f).
In one particularly preferred embodiment, the recombinant glycoprotein is an
antibody and the variable
region of the antibody binds to the affinity ligand immobilized on the solid
support, preferably the affinity
ligand is an antigen. In a preferred embodiment the antibody is an IgG
antibody selected from IgG1,
IgG2, IgG3 or IgG4, more preferably an IgG1 or IgG2 antibody. The antibody may
be a chimeric, human
or humanized IgG antibody, preferably the antibody is a human or humanized
IgG1 or IgG2 antibody.
In another particularly preferred embodiment the recombinant glycoprotein is
an Fc-fusion protein
comprising an Fc-domain and an effector domain and the effector domain binds
to the affinity ligand
immobilized on the solid support, wherein the affinity ligand is a binding
partner or an antibody
specifically binding to the effector domain of the Fc-fusion protein.
When the glycans of the second glycan containing fragment of the recombinant
glycoprotein of each of
said samples immobilized to the solid support are further analyzed, a pre-
clearing step of the liquid
sample is required. Preferably, the pre-clearing step is a second affinity
chromatography upstream of
the immobilization of the recombinant glycoprotein in step b), enriching the
recombinant glycoprotein in
the sample following elution. Preferably, recovery rate of the recombinant
protein of interest is high,
e.g., 90% or more, 95% or more, 98% or more, 99% or more or 100 %. For Fc-
domain containing
glycoproteins, an Fc-binding protein may be used for enriching the recombinant
glycoprotein. For Fc-
domain containing glycoproteins the pre-clearing step preferably comprises
aai) immobilizing Fc-
containing glycoproteins on a solid support using an Fc-binding protein,
wherein the solid support is
made of the same material as the solid support used for immobilizing the
recombinant glycoprotein in
step b) without the coupled affinity ligand used in step b) and aaii) eluting
the Fc-containing
glycoproteins. Preferably the Fc-binding protein is protein G or protein A,
more preferably said Fc-
binding protein is protein G. After binding to the Fc-binding protein and
intensive washing, IgGs or Fc-
fusion proteins are eluted. Suitable methods to elute IgGs and fusion proteins
from Fc-binding proteins
are known to the person skilled in the art, and include without being limited
thereto the use of acidic
conditions, such as glycine or arginine buffer of low pH. Due to the typically
lower affinity of protein G
(or protein A) to the Fc-domain compared to the binding of an antibody to its
antigen or of a therapeutic
fusion protein to its target, the recovery is almost complete. Non-
specifically bound endogenous
glycoproteins that could falsify the analysis remain bound to the solid
support. The pre-cleared sample
is neutralized and added to an affinity column with immobilized affinity
ligand.
24
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
For Fc-fusion proteins, following immobilization and release of the Fc-fusion
protein by enzymatic
cleavage (steps b and c), the glycans of the effector domain of the Fc-fusion
protein remaining on the
solid support, are released into a solution. The glycans are labeled with a
fluorescent label following
glycan release and analyzed together with the reference standard. The
reference standard comprising
glycans labeled with a second stable isotope of the fluorescent label may be
added on-column prior to
release of the glycans of the effector domain (step dd), prior to the labeling
of the glycans to be
analyzed (step ee) or prior to analyzing the labeled glycans using LC-MS (step
if).
Preferably the reference standard comprises N-glycans from a mixture of
effector domain and Fc-
domain glycans labeled with a stable isotope variant of the fluorescent
labeled, such as a stable isotope
variant of 2-AA.
In one embodiment the recombinant protein is an antibody or an Fc-fusion
protein and the pre-clearing
step comprises aai) immobilizing the Fc-containing glycoproteins on a solid
support using an Fc-binding
protein, wherein said Fc-binding protein is preferably selected from protein G
or protein A, more
preferably said Fc-binding protein is protein G; and aaii) eluting the Fc-
containing glycoproteins;
wherein the solid support used in said pre-clearing step is made of the same
material as the solid
support used in step b), but is not coupled to said affinity ligand.
The enzymatic cleavage according to the methods of the invention, releasing
the glycan containing
fragment of the glycoprotein may be done using an endoglycanase or an
endoproteinase, wherein
endoproteinases are preferred. Preferably the endoproteinase cleaves the
recombinant glycoprotein
efficiently and specifically (e.g., only at a known cleavage site). More
preferably the enzyme is further
selective for the recombinant glycoprotein of interest. This means that the
specific cleavage site is rare
and only present in the recombinant glycoprotein of interest and few other
glycoproteins, such as a
specific family of glycoproteins, or a group of glycoproteins all comprising
the same domain, such as
Fc-containing glycoproteins. A more selective enzyme avoids or reduces elution
of fragments of non-
specifically bound glycoproteins from the solid support. Preferably, the
released glycan containing
fragment is a peptide containing one glycan or in the case of a released Fc-
domain two peptides
containing one glycan each. For example the endoproteinases suitable in the
methods of the invention
to release the glycan containing Fc-domain of the recombinant glycoprotein
from the solid support are
endoproteinase, such as papain, ficin, cysteine protease SpeB (FabULOUS) or
cysteine proteinase
IdeS (FabRICATORO), preferably the endoproteinase is the cysteine proteinase
IdeS (FabRICATORO).
IdeS specifically cleaves human IgG in the hinge region between the two
glycines of the constant
sequence ELLGGPS. Enzymatic cleavage using endoproteinases is typically very
efficient, resulting in
almost complete release of the glycan containing glycoprotein fragment.
Compared to acidic elution of
therapeutic antibodies or fusion proteins immobilized via their respective
ligand (using for example
glycine of arginine), elution via enzymatic cleavage is often more efficient
and complete and therefore
results in more sensitivity of the method of the invention. Non-selective
endoproteinases, such as Lys-C
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
that hydrolyses specifically at the carbonyl side of Lys or trypsin, cleaving
several times within most
glycoproteins are not well suited in the context of the present invention.
The enzymatic cleavage may also be done using a glycanases, which have
specificity for certain N- or
0-glycosylation sites. Such a glycanase is for example EndoSGlycanase
(IgGZER00), which is
selective for the N-glycan in Fc-domain containing glycoproteins. However, it
cleaves GIcNAc-61,4-
GIcNAc bonds in the N-glycan of the Fc-domain and therefore releases a glycan
lacking fucose. Thus,
EndoSGlycanase is only suitable where fucosylation is not of interest.
In fusion proteins, N-glycans of the effector domain and the fusion partner,
such as an Fc-domain,
albumin or transferrin can be analyzed separately to gain more information
about site specificity. In the
following the method according to the invention will be exemplified for an Fc-
fusion protein. However,
the skilled artisan knows how to adapt the method to other fusion proteins.
For individual analysis of the
Fc part and effector domain N-glycans the two parts have to be separated. The
fusion protein is
therefore immobilized with its effector domain on a solid support, such as a
sepharose resin and the Fc
part is released enzymatically using an enzyme such as IdeS enzyme. IdeS is an
endopeptidase which
selectively cleaves IgGs and related molecules with high specificity below the
hinge region producing a
Fab2 and Fc fragments. IdeS also cleaves IgG Fc-domain containing fusion
proteins. The fusion protein
is cleaved into an effector domain (e.g., receptor part), which is connected
by disulfide bridges and able
to bind the interaction partner or antigen and an Fc-domain. Alternatively any
other endoproteinase
could be used that cleaves selectively between the glycans of the effector
domain and the glycans of
the Fc-domain. Alternatively for other fusion proteins suitable
endoproteinases cleave between the
glycans of the effector domain and the glycans of the fusion partner, such as
albumin or transferrin.
Preferably, the endoproteinase only cuts the fusion protein between the
glycans of the effector domain
and the glycans of the Fc-domain. Even more preferably, the endoproteinase
selectively cuts the fusion
protein or a group (or family of proteins including the fusion protein) within
the sample, but not all
proteins within the sample. This provides further specificity of the method
and reduces contamination
with fragments of non-specifically bound glycoproteins. For example, the IdeS
enzyme cleaves IgG Fc-
domains below the hinge region in antibodies and Fc-fusion proteins and
therefore reduces
contamination of the released Fc-domain by other contaminating glycoproteins
that do not have an Fc-
domain. The fusion protein may also contain a specific cleavage site, which
has been introduced by
molecular cloning between the effector domain and the fusion partner, which
may be used in this
method.
The fluorescent label as used in the methods of the invention may be any label
suitable for LC-MS
analysis such as 2-amino benzamide (2-AB), 2-amino benzoic acid (2-AA) or 8-
aminonachthalene-1, 3,
6-trisulfonic acid (ANTS). However, there must be a stable light and a stable
heavy isotope variant
available to label the glycans to be analyzed and the reference standard with
a different isotope variant
of the same fluorescent label. Preferably, the fluorescent label is a 12[C]
and 13[C] isotopic pair, more
26
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
preferably a 12[¨
U6] and 13[C6] isotopic pair, and even more preferably a 12[C6]-2-aminobenzoic
acid
(12-6,-2
[u j -AA) and a 13[C6]-2-aminobenzoic acid (13[C6]-2-AA) isotopic pair.
It will be appreciated that the labeling with 2-AA in accordance with the
methods of the present
invention is particularly advantageous for mapping (i.e., analyzing) neutral
or acidic glycans of a
recombinant glycoprotein, such as an antibody or a fusion protein. The glycans
to be analyzed are
preferably N-glycans, particularly N-glycans that are either of the
oligomannose, hybrid or complex type.
In one embodiment the N-glycans are G1F isomers, particularly the G1F isomers
with 1,3 or 1,6
galactosylation.
Following fluorescent labeling of the glycans in the methods of the present
invention, access label may
be separated using gel filtration. For example Sephadex G10 columns may be
equilibrated with water
and the mixture comprising fluorescently labeled glycans, free fluorescent
label and optionally the
reference standard is added to the gel filtration column. The labeled glycans
and the reference standard
are eluted with water following by evaporating the water in a vacuum
centrifuge and the dried sample
may be dissolved in water for LC-MS analysis. The skilled person will
understand that the gel filtration
can be performed with many samples in parallel by using multiwell plates such
as 96 well plates or
other sizes.
Release of the glycans from the glycan containing fragment of the recombinant
glycoprotein can be
done chemically or enzymatically by methods well known to those skilled in the
art.
For example, chemical release of glycans for subsequent labeling (or analysis,
as the case may be)
may be affected by hydrazinolysis or by alkaline 13-elimination, methods that
are well known to those of
skill in the art. An exemplary useful kit for chemical removal is the
GlycoProfileTM IV Chemical
Deglycosylation Kit offered by Sigma-Aldrich.
Release of the glycans by enzymatic methods is preferred and causes no
problems to those of skill in
the art. A particularly preferred method of glycan removal for subsequent
labeling (or analysis) is
digestion with PNGaseF (Peptide N-glycosidase F). Various suitable PNGaseF
enzymes are offered
commercially under different trade names (e.g., N-Glycanase ), which are
mostly engineered or
optimized PNGaseF, although using an engineered or optimized PNGaseF is not
mandatory in the
context of the present invention. Enzymatic release may also be done by using
Endoglycosidase H (or
an enzyme with similar enzymatic activity, such as EndoSGlycanase IgGZEROTM)
or Endoglycosidase
F2, which cleave between the two N-Acetylglucosamines of the glycan core
leaving the first
monosaccharide attached to the protein. However, the information whether the
glycan carried a fucose
or not is lost in this way. Furthermore Endoglycosidase H and Endoglycosidase
F2 are only specific for
oligomannose and hybrid bi-antennary glycans.
For glycan analysis in liquid samples of a mammal, such as serum, on-column
deglycosylation should
only be used in combination with a pre-clearing step prior to step b), because
endogenous serum
glycoproteins that bind non-specifically to the sepharose resin and cannot be
removed by additional
27
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
washing steps. With only one N-glycosylation site at the Fc part, antibodies
are particularly suited for an
enzymatic elution step. The enzyme IdeS selectively cleaves IgG heavy chains C-
terminal of the
disulfide connection thereby releasing the two glycosylated heavy chain
fragments that can
subsequently be deglycosylated with high efficiency. Theoretically more than
one enzymatic elution
step can be performed after one another. For glycan analysis of the
immobilized glycan containing
fragment of the recombinant glycoproteins, on-column deglycosylation is
performed in combination with
a pre-clearing step prior to immobilization of the recombinant glycoprotein to
the solid support (step b)).
Preferably, the pre-clearing step comprises affinity purification to enrich
the recombinant glycoprotein in
the sample.
The LC-MS used in the methods of the present invention is preferably a reverse
phase LC-MS or a
NanoLC-MS, more preferably the LC-MS is a reverse phase NanoLC-MS.
Methods for affinity purifying antibodies, fusion proteins or other
glycoproteins are known to the skilled
person. The solid support used may be a resin, such as a resin comprising
microbeads, preferably
sepharose beads, agarose beads or magnetic beads, more preferably sepharose
beads. However any
solid support or resin suitable for affinity chromatography can be used in the
methods of the present
invention. The affinity ligand may be coupled to the solid support via N-
hydroxysuccinimide (NHS),
cyanogen bromide, epoxy, carbodiimide or thiopropyl, preferably via N-
hydroxysuccinimide (NHS).
Again, methods for coupling affinity ligands to a solid support are known in
the art. Typically resins
activated with one of the above linkers are commercially available.
The recombinant glycoproteins that will typically be analyzed using the
methods of the present invention
are therapeutically relevant recombinant glycoproteins. Therapeutically
relevant recombinant
glycoproteins often have a very high affinity and specificity to its
therapeutic target. Hence the
therapeutic target is well suited as an affinity ligand to capture the
recombinant glycoprotein from the
liquid sample of a mammal and immobilize it to the solid support. For example
affinity purification of an
antibody with the respective antigen has the advantage of very high affinity
and specificity due to the
strong interaction of the antibody with its antigen. The affinity of a
therapeutic antibody for its antigen is
typically in the picomolar to low nanomolar range. Therapeutically relevant
fusion proteins, such as
etanercept, typically bind their therapeutic target with a similar affinity.
However, one major drawback of
this high affinity binding is the limited recovery of antibodies or fusion
proteins after acidic elution. Thus,
by eluting a glycan containing fragment of the glycoprotein using enzymatic
cleavage the recovery may
be improved.
The affinity ligand may be any binding partner, including without being
limited thereto the ligand for a
receptor, the substrate, an antigen or an antibody. For therapeutically
relevant glycoproteins the affinity
ligand is preferably the therapeutic target or a fragment thereof. The
affinity ligand may also be a
mutant of the therapeutic target or a fragment thereof, preferably a mutant
binding to the recombinant
glycoprotein with higher affinity compared to the wild type therapeutic target
or fragment thereof. Where
the recombinant glycoprotein is an antibody, the affinity ligand is preferably
an antigen binding to the
28
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
antibody, more preferably an antigen binding to the antibody with high
affinity. The antigen may be for
example a complex of more than one proteins, a full length protein or a
fragment thereof, including a
short peptide. For fusion proteins the affinity ligand may be a binding
partner preferably binding to the
effector domain of the fusion protein. The affinity ligand may also be an
antibody binding to the effector
domain of a fusion protein. For any other glycoprotein of interest, the
affinity ligand may likewise be a
binding partner or an antibody binding to the recombinant glycoprotein of
interest. Preferably, the affinity
ligand is not glycosylated.
In a preferred embodiment of the methods of the invention the two or more
liquid samples of a mammal
or aliquots thereof are prepared for analysis using a multi well filter plate,
preferably a 24, 96 or 384 well
filter plate, more preferably a 96 well filter plate. The filter membrane is
preferably a low protein binding
and/or hydrophile membrane, such as nitrocellulose or polyvinylidene
difluoride (PVDF) membranes.
These filter plates are suitable for affinity chromatography, enzymatic
digest, fluorescent labeling and
gel-filtration, depending on the reagents or resins added.
The liquid sample of a mammal comprising the recombinant protein is a body
fluid and is preferably
selected from the group consisting of serum or plasma, urine, cerebral spinal
fluid, amniotic fluid, saliva,
sweat, ejaculate, tears, phlegm, vaginal secretion, vaginal wash and colonic
wash; preferably said
sample is a plasma or a serum sample. Preferably the samples to be analyzed
have a volume from
about 1 pl to about 1000 pl, from about 5 pl to about 500 pl, from about 10 pl
to about 200 pl, from
about 10 pl to about 100 pl, from about 25 pl to about 100 pl, or from about
40 pl to about 75 pl,
preferably of about 50 pl. In a preferred embodiment the two or more liquid
samples of a mammal were
obtained from the same subject. The mammalian liquid sample analyzed in the
methods of the
invention may be a human, a monkey, a rodent, a dog, a cat or a pig sample,
preferably a rodent
sample such as from mouse, rat, hamster or rabbit.
The methods of the present invention may be used to determine pharmacokinetic
parameters of at least
one specific glycan structure of the glycans of the recombinant glycoprotein
of interest, preferably the
Cmax, tmax, AUG or t112. Preferably the two or more liquid samples of a mammal
were obtained at different
time points from the same subject, wherein at least about 24 h, 48 h, 72 h, 96
h, 120 h, 144 h, 168 h,
336 h, 504 h, 672 h, 840 h may be between the first and the last time point.
Preferably the two or more
liquid samples of a mammal are 5 or more, 10 or more, 20 or more, 30 or more,
40 or more or 50 or
more samples.
The methods of the invention may further comprise the steps of (i) determining
the area under the curve
of the extracted ion chromatograms (EICs) of the glycans from each of said
samples labeled with the
first stable isotope of the fluorescent label and of its respective reference
standard glycan labeled with
the second isotope of the fluorescently label to form a ratio of first isotope
sample glycan to second
isotope reference standard over time, and (ii) optionally calculating glycan
percentages based on the
known relative distribution of the individual glycans in the reference
standard for each of said samples.
29
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Further, the ratios or the glycan percentages of the two or more samples for
each individual glycan may
be compared, e.g., plotted against time. Preferably the glycans are N-glycans.
For example the L/H ratios of the sample N-glycans (light signal) to the
constant heavy isotope standard
N-glycans (heavy signal) plotted against time result in glycan PK profiles for
each N-glycan.
Alternatively calculated sample N-glycan percentages may be plotted against
time, resulting in glycan
PK profiles of each N-glycan. Determined glycan PK profiles can be compared
against the respective
ELISA profile and are the basis for calculation of the glycan maps. The
skilled person will understand
that the stable light isotope could equally be used as a label in the
reference standard, when the stable
heavy isotope is used for labeling the glycans to be analyzed. In other words,
as long as the glycans of
the reference standard are labeled with a fluorescent label of a different
isotype variant as the glycans
to be analyzed, the stable light and the heavy isotope fluorescent label are
equally suited. The skilled
person will understand that the percentages (%) of the glycans refer to mol,
because the "heavy" glycan
reference standard composition is determined using the UV signal, which
depends solely on the label
which is directly proportional to the number of labeled glycan molecules.
The glycans may also be analyzed according to the methods of the invention in
an aliquot of each of
said two or more liquid samples of a mammal. Preferably the aliquots to be
analyzed have a volume
from about 1 pl to about 1000 pl, from about 5 pl to about 500 pl, from about
10 pl to about 200 pl, from
about 10 pl to about 100 pl, from about 25 pl to about 100 pl, or from about
40 pl to about 75 pl,
preferably of about 50 pl. Optionally, the concentration of said recombinant
glycoprotein may be
analyzed in a further aliquot of said two or more liquid samples of a mammal.
In one embodiment, the
concentration of said recombinant glycoprotein is analyzed by ELISA or any
other method known to the
person skilled in the art to determine concentrations of a specific protein in
a sample.
The liquid chromatography-MS to be performed in connection with the methods of
the present invention
is preferably reverse phase liquid chromatography (RP-LC-MS), more preferably
reversed-phase high
performance liquid chromatography (RP-HPLC-MS), or ultra performance liquid
chromatography
(UPLC-MS). In this case, as in other preferred cases of RP-LC performed in
connection with the
present invention, these chromatography methods are performed on a reversed-
phase liquid
chromatography column. Preferably the RP-LC is performed under conditions
under which the carboxy
group of the 2-AA label attached to the glycans is neutral. Reverse phase LC
is preferred in the
methods of the present invention as it allows more sensitive analysis compared
to normal phase LC,
because the mobile phase is more compatible with mass spectrometry.
Optionally, no ion-pairing
reagent is used in the mobile phase. Preferably, an acidic mobile phase is
used for the RP-LC.
The liquid chromatography to be performed in connection with the methods of
the present invention is
preferably a nano-LC or nano-HPLC, even more preferably a reverse phase nano-
LC or a reverse
phase nano-HPLC. Nano-LC is characterized by a decreased inner diameter of the
columns that are
used for LC (10-150 pm) and smaller flow-rates (10-1000 nl/min) compared to
conventional LC or
HPLC, respectively. This down-scaling results in the ability to analyze
proteinaceous samples in the low
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
femtomole and subfemtomole ranges (Chervet et al., Analytical Chemistry 1996,
68:1507-12) and
therefore allows to reduce the sample volume to be analyzed and the duration
to detect glycans of
recombinant glycoproteins post administration
A suitable mobile phase used during RP-LC comprises formic acid. In this
regard, preferred amounts of
formic acid in the mobile phases are from about 0.1% to about 2.0% formic
acid. Typical preferred
amounts therefore include about 0.2%, about 0.3%, about 0.4%, about 0.5%,
about 0.6%, about 0.7%,
about 0.8%, about 0.9%, about 1.0%, about 1.1%, about 1.2%, about 1.3%, about
1.4%, about 1.5%,
about 1.6%, about 1.7%, about 1.8%, or about 1.9% formic acid. An amount of
about 1.0% formic acid
in the mobile phase is particularly preferred.
Another suitable mobile phase used during RP-LC comprises acetic acid. In this
regard, preferred
amounts of acetic acid in the mobile phases are again from about 0.1% to about
2.0% acetic acid.
Typical preferred amounts therefore include about 0.2%, about 0.3%, about
0.4%, about 0.5%, about
0.6%, about 0.7%, about 0.8%, about 0.9%, about 1.0%, about 1.1%, about 1.2%,
about 1.3%, about
1.4%, about 1.5%, about 1.6%, about 1.7%, about 1.8%, or about 1.9% acetic
acid. An amount of about
1.0% acetic acid in the mobile phase is particularly preferred.
Further, suitable pH values of the mobile phase used during RP-LC are in the
range of about 1 to about
4, more preferably in the range of about 1.5 to about 3, yet more preferably
of about 1.8 to about 2.9,
even more preferably of about 1.9 to about 2.75, and particularly preferably
of about 2 to about 2.7.
Preferred is in particular a pH value in the range of about 2.1 to about 2.18.
Accordingly, preferred
mobile phases used during RP-LC in accordance with the methods and uses
described and/or claimed
herein will have a pH value of about 1.9, 2, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, or
2.7, with pH values of about
2.1 or of about 2.18 being particularly preferred.
In the course of the methods of the present invention the separation of the
fluorescent (preferably 2-AA)
labeled glycans by LC or RP-LC may be advantageously performed at temperatures
in the range of
about 4 C to about room temperature, or at room temperature (with room
temperature being defined as
23 C for the purpose of the present invention). Preferably, however, the
separation is performed at
temperatures above room temperature. Preferred in this regard is a temperature
in the range of about
40 C to about 60 C, with about 50 C being particularly preferred.
Suitable flow rates for a nanoRP-LC or nanoLC column in accordance with the
present invention will be
readily known or determined by those of skill in the art. Generally, suitable
flow rates will typically be in
a range of about 50-1000 nl per minute. Preferred are, for example, flow rates
in a range of about 100-
500 nl per minute. Accordingly, preferred values for the flow rate are about
100 nl per minute, about 200
nl per minute, about 300 nl per minute, about 400 nl per minute, or about 500
nl per minute, with a flow
rate of about 300 nl per minute being particularly preferred.
As noted above, the methods of the present invention comprise subjecting the
fluorescently labeled
glycans (preferably 2-AA labeled glycans) to mass spectrometry (MS) after
their separation via LC or
31
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
RP-LC, with methods where this is done by directly coupling LC or RP-LC with
the MS analysis being
particularly preferred. Mass spectrometry methods suitable in this regard
include ion-trap mass
spectrometry, such as positive ionization mass spectrometry.
In a preferred embodiment the methods of the invention allow for analyzing
attomolar concentrations of
the individual labeled glycans, e.g., concentrations as low as 800 amol,
preferably as low as 600 amol,
and more preferably as low as 400 amol. In another preferred embodiment the
method allow for
analyzing the glycans of the recombinant glycoprotein at a recombinant
glycoprotein concentration in
each of the two or more liquid samples of a mammal of 20 pg/ml or less, 10
pg/ml or less, 5 pg/ml or
less, 2 pg/ml or less, 1 pg/ml or less, 0.5 pg/ml or less, 0.2 pg/ml or less
or 0.1 pg/ml or less. In yet
another preferred embodiment the methods of the present invention allow for
analyzing the glycans of
the recombinant glycoprotein in in the two or more liquid samples of a mammal
comprising 1.0pg or
less, 0.5 pg or less, 0.25 pg or less, 0.1 pg or less, 0.05 pg or less, 0.025
pg or less, 0.01 pg or less, or
0.005 pg or less of the recombinant protein each. In yet another preferred
embodiment at least steps b),
c), d) and e) and/or at least steps aa), dd) and ee) are operated in a high
throughput manner.
The present invention also relates to a method of preparing a glycoprotein
based pharmaceutical
composition comprising analyzing the glycans of a recombinant glycoprotein in
liquid samples of a
mammal of according to any one of the preceding claims; and formulating the
glycoprotein into said
pharmaceutical composition. Optionally the recombinant glycoprotein is
glycoengineered prior to
formulating the glycoprotein into the pharmaceutical composition.
All publications as cited herein are incorporated herein by reference in their
entirety.
The invention is further illustrated by the following Figures and Examples,
which are not to be
considered as being limiting for the scope of protection conferred by the
claims of the present
application.
Examples
Materials
2-Aminobenzoic acid, ethanolamine, formic acid, picoline borane, DMSO, 13C
aminobenzoic acid were
from Sigma (Munich, Germany). PNGaseF was from Roche (Penzberg, Germany).
Acetic acid,
acetonitrile and hydrochloric acid were from Merck (Darmstadt, Germany).
Protein G, NHS activated
sepharose, Sephadex G-10 96-well plates and 96-well deep well plates were
from GE Healthcare
(Munich, Germany). TNF-alpha was from Peprotech (Hamburg, Germany). Phosphate
buffered saline
was from Gibco/Life technologies (Darmstadt, Germany).
Multicreen THS HV filter plates were from Milipore. Fabricator was Genovis
(Lund, Sweden). 96-well
plates were from Nunc/Thermo Scientific (Munich, Germany). AcroPrepTM Advance
OmegaTM 10K 96-
well filter plates were from Pall (Dreieich, Germany). Pre-clinical rabbit
serum samples were obtained
from clinical bioanalytics at Sandoz.
32
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Example 1: Influence of glycol-variants on the pharmacokinetics of an IgG1
biopharmaceutical
Preclinical rabbit study
The preclinical study was performed in New Zealand White rabbits. Following
single subcutaneous
administration of 10 mg kg-1 body weight of an IgG1 mAb1 (anti-TNF-a antibody)
blood samples were
drawn over a period of time including one pre-dose blood sample. Serum samples
were taken at 12
time points after administration. Detailed sampling is listed in Table 1.
Concentration of mAb1 in serum
was determined by ELISA. From remaining serum 2 x 50 pl aliquots were used for
glycan PK profiling.
The first aliquot was analyzed and the second aliquot served as back-up
aliquot.
Table 1: Sampling schedule of the pre-clinical study of an IgG1. At each
sampling time point ¨500 pl of
serum were drawn.
Day 1 1 1 2 2 3 3 4 5 8 15 22 29
Hours post- 0
2 8 24 40 48 60 72 96 168 336 504 672
dose (pre-dose)
Reconstitution of the antigen
Recombinant human antigen (TNF-a) produced in E.coli was reconstituted
according to the
manufacturer instructions. Antigen was dissolved in H20 (1 mg/mL) and
reconstituted for 2 hours at
room temperature.
Preparation of 13C 2-AA labeled glycan standard
N-glycans of desalted mAb (1 mg) were released using PNGaseF digest overnight
(17 h) at 37 C. The
N-glycans were separated from the proteins by use of Amicon 30K filter devices
and were brought to
dryness using a speedvac. Picoline borane and [13C] 2-AA were dissolved in
70:30(% v/v) DMSO-acetic
acid to furnish concentrations of 63 and 50 mg/ml, respectively. Labeling
solution (15 pL) and deionized
water (10 pL) were added to 15 nmol enzymatically released and dried glycans.
The labeling reaction
was performed at 37 C for 17 h.
Excess label was removed by gel filtration on G-10 columns. Columns were
conditioned with 10 ml
H20. Samples were diluted to 100 pl with deionized water and applied to the
column. After rinsing the
column with 700 pl H20 the purified fluorescence labeled N-glycans were eluted
with 600 pl H20.
Purified [13C] 2-AA labeled N-glycans are aliquoted and stored at -20 C until
use.
Preparation of 96-well plate affinity columns with immobilized antigen
The membranes of a 96 well filter plate comprising a hydrophilic low-protein
binding membrane
(Multiscreen THS HV filter plates) were wetted with 1mM HCI (100 pL) before
addition of 200 pL NHS
activated sepharose-isopropanol slurry per well. Isopropanol was removed by
centrifugation and the
columns were washed with 1 mM HCI (150 pL) for four times. Antigen solution
(100 pL; 50 p1/m1) was
centrifuged into the columns and coupling reaction was allowed to take place
for 2 hours at ambient
33
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
temperature. Affinity columns were washed and remaining NHS groups were
inactivated using
ethanolamine buffer (150 pL). Finally, columns were equilibrated with PBS.
Affinity purification of an IgG1 biopharmaceutical and glycan release
Serum samples (50 pL) were diluted to 100pL with PBS and applied to the
affinity purification column
by centrifugation. Bound mAb was washed several times with PBS to remove serum
and non-specific
bound proteins. Fabricator solution (1 U/pl) was centrifuged into the columns
to release the glycosylated
Fc part of the mAb. Reaction was performed at 37 C for 30 minutes. Released Fc
parts were eluted
with PBS and PNGaseF with 13C-2-AA labeled N-glycan standards was added. This
mixture was
incubated for 17 hours at 37 C. Remaining proteins were removed by
ultrafiltration using 96-well plates
with 10 K cut-off membranes. Released N-glycans with glycan standard were
dried by vacuum
centrifugation.
N-glycan labeling
Dried samples containing free reducing end N-glycans and 13C 2-AA labeled
glycan standard were
dissolved in H20 (10 pL), 2-AA labeling solution (15 pL; 100 mg/mL picoline
borane, 50 mg/mL 2-AA in
7:3 mixture of DMSO and acetic acid) was added and incubated for 17 hours at
37 C.
Gel filtration
Excess label is subsequently removed by gel filtration. Custom made 96-well
plate Sephadex G-10
columns were equilibrated with 800 pL H20. Labeled samples were filled up to
100 pL with H20 and
applied to the gel filtration columns. 2-AA and 13C 2-AA labeled N-glycans
were eluted with H20 (150
pL). Finally samples were brought to dryness by vacuum centrifugation and were
dissolved in 20 pL
H20 for nanoLC-MS analysis.
NanoLC of labeled N-glycans
Labeled sample N-glycans and the heavy isotope standard were analyzed by RP
nanoLC-MS. NanoLC
(Thermo/Dionex Ultimate 3000) was set-up in "pre-concentration" mode according
to the manufacturer
manual with a pre-concentration column (3 pm particles, 75 pm x 2 cm) and an
analytical column (2 pm
particles, 75 pm x 25 cm). Column compartment was held at 40 C. Mobile phase
of the nano pump
consisted of 0.5% formic acid in H20 (component A) and 0.5% formic acid in 50%
ACN (component B).
Mobile phase of the capillary pump consisted of 0.5% formic acid and 1% ACN in
H20 (component C).
The analytical column was equilibrated with 2% component B at a flow rate of
300 nl/min. Pre-
concentration column was equilibrated with 100% component C. With a user
defined injection routine, 8
pl sample were stacked between loading solution (0.1% formic acid, 1% ACN in
ultrapure water) in a 20
pl sample loop. Sample loop was switched for 2 minutes in-line of the
capillary pump flow to allow
optimal trapping. Prior to the next injection sample the loop was washed with
loading solution. After
trapping the pre-concentration column was switched into the nano pump flow and
component B was
raised to 30% over 60 minutes, then to 95% over 5 minutes. After holding at
95% component B for 5
34
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
minutes the column was finally re-equilibrated at 2% component B for 15
minutes. Column outlet was
connected to a UV detector with a 3 nl flow-cell.
Mass Spectrometry
The outlet of the nanoLC was directly coupled to an ion trap ESI-MS (Bruker
AmaZon) equipped with an
on-line nano source (Bruker CaptiveSpray ). The ion trap was operated in
Enhanced Resolution Mode
with a capillary voltage of 1.7 kV. Source temperature was set to 200 C and a
dry gas flow of 3 l/min
was used to heat the source.
Data interpretation is done by determining the area of the respective light
sample glycans and its
appropriate heavy isotope standard from their EICs. This light to heavy ratio
(L/H) is determined for
each N-glycan at each time point and animal. By plotting the L/H ratio against
time PK profiles for each
N-glycan are obtained. The N-glycan percentages can be calculated with the
known relative distribution
of heavy isotope standard glycans for each time point.
mAb1 N-glycosylation and qualification of the study
To qualify and control the affinity purification, QC samples were prepared by
spiking known amounts of
mAb1 into NZW rabbit serum. The QC samples covered the concentration range of
the study which was
previously determined by ELISA. Figure 2A shows the mean percentages of the
mAb1 QC samples.
The samples covered the concentrations between 10 pg/ml and 100 pg/ml. In
total four QC samples (0
pg/ml, 1 pg/ml; 10 pg/ml and 100 pg/ml) in duplicates were used including one
serum blank. The mean
percentages of the glycans obtained from the QC samples represent the mAb1 N-
glycan composition.
With the developed enzymatically elution of the glycosylated antibody
fragments very high selectivity
and purity could be achieved. No interfering N-glycans of mAb1 were co-
purified with the exception of
two minor abundant bisecting variants that were excluded from analysis. No
additional serum related N-
glycans were detected. Sensitivity was at least sufficient to analyze all N-
glycans with a percentage of
at least 0.1 % at a concentration of 10 pg/ml (LLOQ). However, sensitivity can
be improved to about 0.1
p/ml using improved nano ESI sources.
The N-glycans was mainly complex biantennary with core fucose (Figure 2B). As
may be taken from
Figure 2A, the most abundant N-glycan was GOF (65 %) with terminal N-
acetylglucosamine residues
followed by G1F (16 %) with one additional terminal galactose. High mannose
glycan M5 (9.5 %) was
the third most abundant N-glycan. All other N-glycans had a percentage of less
than 3%. mAb1
contained no N-glycans with terminal sialic acids.
Preclinical study: Glycan PK profiles of mAb1 and ELISA
The determined glycan L/H ratios were plotted against the sampling time to
obtain PK profiles for each
N-glycan. Mean L/H ratios for each N-glycan were normalized to the maximum.
ELISA data were
normalized in the same way. The relative concentrations of the ELISA and L/H
values were then
compared graphically. Figure 3 shows the resulting curves of the N-glycans.
Mean ELISA profiles are
depicted in diamonds with error bars representing the variability. The most
abundant complex type GOF
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
had a very similar PK profile compared to the ELISA profile (Figure 3A). The
maximal concentration
(tmax) was achieved after 72 h with congruency of the two graphs. At the
beginning of elimination there
was a slight difference, however, with complete overlap of the error bars the
PK profiles can be
regarded comparable. This finding was expected for the major N-glycan with
almost 70 % relative
abundance, as the ELISA represents the average profile of all glycoforms to
which the most abundant
N-glycan contributes most.
The second most abundant complex type N-glycan G1F that accounts for
approximately 16 % showed
also a very similar PK profile when compared to the ELISA profile (Figure 3B).
Again tmax was reached
at 72 h for both curves.
Complex type G2F with a percentage of 2 % had the best match with the ELISA
profile showing almost
perfect congruency (Figure 3C). tmax was reached earlier with 60 h compared to
72 h of the ELISA
curve. Due to the very small difference between the 60 h and 72 h time point
and the fact that there was
no additional sampling point, the two profiles can still be regarded
identical.
The profile of the M6G0F/M5G1F hybrid type glycan, which are two isomers that
could not be
differentiated with the nanoLC-MS approach and had a relative content of only
0.1 %, still looks similar
to the ELISA profile (Figure 3D). The low abundance was close to the LLOQ
which resulted in lower
precision of the L/H values than for the other more abundant glycans. However,
from the similar tmax
and the similar curves it can be concluded that the PK was identical or at
least similar.
These results demonstrate that PK profiles can be obtained for each N-glycan
individually. The glycan
PK profiles were highly similar to the ELISA for the most abundant glycans.
Merely for N-glycans with a
portion smaller than 0.5 % the graphs were not that clearly due to the higher
variation as a result of the
small abundance.
The glycan PK profiles of the high mannose type N-glycans M5 and M6 showed a
completely different
picture. The curves of the N-glycan with 9.5 % and 2 % relative abundance for
M5 and M6, respectively
are shown in Figure 4. M6 had a PK profile that deviated extremely from the
average profile determined
by ELISA (Figure 4A). The maximum concentration was achieved after 24 h
followed by either a
conversion to M5 or an increased elimination rate that lead to almost complete
removal from circulation.
High mannose glycan M5 also had a different PK profile when compared to ELISA
(Figure 4B). tmax was
reached 24 h earlier after 48 h and furthermore clearance was faster between
48 h and 168 h. These
results demonstrate that high mannose glycan PK profiles of mAb1 were
different in comparison to the
ELISA which represents the average concentration of all protein variants
including glyco-variants.
These findings are also reflected in the glycan maps that were determined for
each time point (see
Figure 5) to strengthen the theory of a conversion and increased clearance of
high mannose glycans.
PK profiles of M3G1F, M3GOF and the core structure M3 are shown in (Figure 3E-
G). PK profiles for GO
could not be obtained due to co-elution of a contaminant with the same m/z
value.
36
CA 02954364 2017-01-05
WO 2016/009077
PCT/EP2015/066488
Selective clearance of M5 and M6
The composition of mAb1 N-glycosylation is shown in Figure 2. Based on the
known relative amount of
the heavy isotope glycan standard and the experimentally obtained L/H ratios
the percentage of each
N-glycan was calculated. Figure 5 shows the resulting glycan maps of the
preclinical study. Glycan
maps were calculated for time points between 8 h and 336 h. At 2 h, 504 h and
672 h the serum
concentration of mAb1 was below 10 ug/m1 for many rabbits, which is the LLOQ
of this study, and
therefore no glycan maps could be calculated.
Mean percentages of the most abundant N-glycans GOF and G1F stayed constant,
which confirm
previously observations (Figure 5 left). The magnified view shows the minor
abundant glycans (Figure 5
right). The percentages stayed constant for all N-glycans except the high
mannose glycans. M5 and M6
portion decreased over time. M6 was removed from circulation whereas the M5
portion decreased from
the initial percentage of 9 % to approximately 4 %.
It is known from literature that high mannose species M6-M9 are converted to
smaller high mannose
glycans like M5 in humans (Chen et al., Glycobiology (2009) 21, 949-59;
Alessandri et al., MAbs (2012)
4(4), 509-520). In mice a similar observation was made. High mannose species
M7-M9 are converted
to M6 (Yu et al., mAbs (2012) 4(4), 475-87). Both observations were made for
in vitro incubation of
antibody in serum. In the present study mainly M5 was removed selectively from
circulation. An
increase of smaller glycans like M4 or M3 (core structure) was not observed. A
further conversion of M5
would require additional enzymes with other specificities in serum, because
the linkage of the terminal
mannose residues of M5 is different to that of M6-M9. It is very likely that
M5 or rather M5 containing
glycoproteins, are removed by a specific clearance mechanism from circulation
e.g. through the
mannose receptor.
Table 2: Glycoforms containing M5.
M5:M5 M5:G1F M5:M3G1F M5:GOF M5:M3GOF GOF:GOF
Calculated
0.62% 1.12% 0.11% 5.29% 0.18% 45.43%
relative*
Calculated
relative to 1.36% 11.64%
100%
GOF:GOF
Observed
relative to 5.13% 12.73%
100%
GOF:GOF
*Values are calculated based on the assumption of random pairing.
The incomplete removal of M5 from circulation can be explained with the
structural conformation of the
Fc part. Several investigations showed that IgG glycoforms containing N-
glycans with terminal
galactosylation, which have a similar size as M5, change the conformation of
the Fc part to an open
conformation, a horseshoe conformation which makes the N-glycans accessible
for other proteins to
bind (Krapp et al., J. Mol. Biol. (2003), 325, 979-989). This implies that M5
pairs with N-glycans
37
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
resulting in a glycoform that is large enough to force the Fc part into an
open conformation. It was
shown that pairing of the glycosylated heavy chains during protein
biosynthesis is not random (Masuda
et al., FEBS Lett. (2000), 473, 349.57. For an IgG2 biopharmaceutical it was
demonstrated that
glycoform M5:M5 is favored (Goetze et al., Glycobiology (2009) 19, 240-9). For
mAb1 in the present
study similar observations were made (Table 2). Assuming random pairing, all
glycoforms can be easily
calculated, which is shown for M5 containing glycoforms. MS data generated
from intact Fc parts of
mAb1 show that the M5:M5 glycoform was four times higher than the calculated
value. M5:M5 was
strongly favored during protein biosynthesis. It is likely that this glycoform
results in a horseshoe
conformation of the Fc part and in turn an increased clearance from
circulation by binding to the
Mannose receptor. The remaining 4 % M5 observed in the preclinical study could
be the M5:GOF
glycoform which would be small enough for a closed Fc conformation. This would
explain the
incompletely clearance from circulation.
Conclusion
A new approach for the investigation of individual N-glycan pharmacokinetics
was utilized to analyze a
preclinical rabbit study of an IgG1 biopharmaceutical. Using 96 well plate
based high throughput affinity
purification with immobilized antigen and stable heavy isotope standard mass
spectrometric
quantification glycan PK data from 50 pl serum samples were obtained for mAb
concentrations between
10 and 90 pg/ml. Glycan PK profiles were compared to ELISA data demonstrating
that high mannose
glycans M5 and M6 had a different PK profile. Glycan maps showed that M6 was
removed during the
first 48 h completely and M5 levels decreased from 9.5 % to approximately 4 %.
The results obtained from the preclinical study presented were similar to the
results that were obtained
from healthy human subjects by another group (Goetze et al., Glycobiology
(2009) 19, 240-9). These
findings demonstrate the value of glycan PK profiling with preclinical
samples. The use of stable heavy
isotope 2-AA glycan standards enables the PK profiling of individual N-glycans
by mass spectrometry
which was so far not described and enables an independent analysis of single N-
glycans in comparison
to so far used relative analysis of the N-glycans.
In summary, this new approach for glycan PK profiling with its high
sensitivity works with preclinical
samples. The results obtained from the preclinical study furthermore
demonstrate that results from
human subjects are identical with findings in preclinical studies.
Implementation during earlier
biopharmaceutical development allows the investigation of N-glyosylation
effects already during
preclinical studies. This allows implementation of preventive actions to
optimize N-glycosylation, glyco-
engineering, before entering clinical phases.
38
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Example 2: Influence of glycol-variants on the pharmacokinetics of a
therapeutic fusion protein
Preclinical rabbit study
The preclinical study was performed in Himalayan rabbits. Following single
subcutaneous
administration of 8 mg kg-1 body weight of two different batches of the Fc-
fusion protein etanercept
(FP1 or FP2), blood samples were drawn over a period of time including one pre-
dose blood sample.
Sampling was performed as listed in Table 3. At each sampling time point at
least 600 pl of serum were
drawn. Concentration of FP1 and FP2 in serum was determined by ELISA. From
remaining serum 2 x
50 pl aliquots were used for glycan PK profiling. The first aliquot was
analyzed subsequently and the
second aliquot served as back-up aliquot.
Preparation of 13C 2-AA labeled glycan standard
N-glycans of desalted FP1 fusion protein (1 mg) were released using PNGaseF
digest overnight (17 h)
at 37 C. The N-glycans were separated from the fusion protein by use of Amicon
30K filter devices and
were brought to dryness using a speedvac. Picoline borane and [13C] 2-AA were
dissolved in 70:30(%
v/v) DMSO-acetic acid to furnish concentrations of 63 and 50 mg/ml,
respectively. Labeling solution (15
pL) and deionized water (10 pL) were added to 15 nmol enzymatically released
and dried glycans. The
labeling reaction was performed at 37 C for 17 h.
Excess label was removed by gel filtration on G-10 columns. Columns were
conditioned with 10 ml
H20. Samples were diluted to 100 pl with deionized water and applied to the
column. After rinsing the
column with 700 pl H20, the purified fluorescence labeled N-glycans were
eluted with 600 pl H20.
Purified [13C] 2-AA labeled N-glycans are aliquoted and stored at -20 C until
use.
Preparation of 96-well plate affinity columns with immobilized Protein G
Protein G sepharose (200 pL) was added to each well of the 96-well filter
plate comprising a hydrophilic
low-protein binding membrane (Multiscreen THS HV filter plates). Protein G
sepharose was stored in
20% ethanol which must be removed before affinity purification. Columns were
therefore equilibrated
with PBS (150 pL) for four times. Liquid was removed by centrifugation.
Preparation of 96-well plate affinity columns with immobilized antigen
The membranes of a 96 well filter plate (Multiscreen THS HV filter plates)
were wetted with 1mM HCI
(100 pL) before addition of 200 pL NHS activated sepharose-isopropanol slurry
per well. Isopropanol
was removed by centrifugation and the columns were washed with 1 mM HCI (150
pL) for four times.
Reconstitution of the antigen expressed in E.coli was performed as described
in Example 1 above.
Antigen solution (100 pL) was centrifuged into the columns and coupling
reaction was allowed to take
place for 2 hours at ambient temperature. Affinity columns were washed and
remaining NHS groups
were inactivated with use of ethanolamine buffer (150 pL). Finally columns
were equilibrated with PBS.
39
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
Affinity purification of a fusion protein and glycan release
Serum samples (50 pL) were added onto the equilibrated Protein G column of the
96 well plate and
centrifuged through the column. The column was subsequently washed with PBS
(150 pL) six times by
centrifugation. Bound IgGs were eluted with three times 100 pL elution buffer
(0.1 M glycine pH 2.7).
Eluat was immediately neutralized with 1 M Tris HCI pH 8Ø The eluat
contained the target protein as
well as other IgGs from serum.
Eluat was added onto the equilibrated immobilized antigen column of the 96
well plate and centrifuged
through the column. Column was then washed six times with PBS (150 pL).
Fabricator solution (100 pl,
1 U/pl) was centrifuged into the columns to release the glycosylated Fc part
of the fusion protein.
Reaction was performed at 37 C for 30 minutes. Released Fc parts were eluted
with PBS. PNGaseF
with 13C-2-AA labeled N-glycan standards was added to the eluted Fc parts and
onto the column with
the antigen bound part. Digests were incubated for 17 hours at 37 C. From
antigen bound protein part
released N-glycans were eluted with H20. Remaining proteins were removed by
ultrafiltration using 96-
well plates with 10K cut-off membranes. Released N-glycans with glycan
standard were dried by
vacuum centrifugation
N-glycan labeling, gel filtration to remove excess 2-AA labeling reagent and
analysis of labeled sample
N-glycans and the heavy isotope standard by RP nanoLC-MS were performed as
described in Example
1.
Glycan maps of the two batches reveal different N-glycosylation
N-glycan PK profiling of monoclonal antibodies is usually less sophisticated
than PK profiling of other
biopharmaceuticals, because mAbs carry only one conserved N-glycosylation site
on each heavy chain
at the Fc part. It is therefore not necessary to analyze the N-glycans with
site specificity. For more
complex glycosylated biopharmaceuticals site specificity provides additional
information that could help
understanding the mechanism of clearance or prolonged half-life. For example
fusion protein consists of
two chains containing the CH2 and CH3 domain of an IgG Fc part and a receptor
part. Like in mAbs the
chains are connected by disulfide bridges. The Fc part carries one N-glycan
per chain and the receptor
part two additional N-glycosylation sites making a total of six N-glycans per
molecule.
For individual analysis of the Fc part and receptor part N-glycans the two
parts were separated using
the endoproteinase IdeS enzyme and the generated fragments were analyzed
individually. After
separation Fc part and receptor part were deglycosylated using PNGaseF. The
glycans were labeled
with 2-AA and analyzed by nanoLCMS. The resulting glycan maps are shown in
Figure 6.
The two fusion protein batches had a different N-glycosylation pattern. The
glycosylation of fusion
protein 1 (FP1) was more heterogenic whereas fusion protein 2 (FP2) has higher
percentage of only
few N-glycans. Both batches contained high percentages of terminal sialic
acids at their receptor part.
The found Fc glycosylation was typical for antibodies with a high portion of
complex bi-antennary N-
glycans and core fucosylation; some carried an additional sialic acid. Main
difference between FP1 and
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
FP2 were the terminal groups. FP1 had a high percentage of terminal GIcNAc
residues. In contrast FP2
had a high degree of terminal galactosylation, while the sialylation was
comparable.
Average PK profiles of FP1 and FP2 in a preclinical study in rabbits
The preclinical study was conducted in Himalayan rabbits. After a single
subcutaneous administration of
FP1 or FP2, samples were taken and the concentration of FP1 and FP2 was
determined by ELISA. The
sampling schedule is depicted in Table 3. Five animals from each arm of the
study were included for
glycan PK profiling.
Table 3: Sampling schedule of the pre-clinical study of FP1 and FP2. Prior to
the administration a pre-
dose sample was taken from each rabbit.
Day 1 1 1 1 2 2 2 3 4 6 8
Hours 0
2 6 12 18 24 32 48 72 120 168
[post-dose] (pre-dose)
The ELISA results of the five animals of both arms are shown in Figure 7. The
PK profiles were
different. FP1 had higher concentrations than FP2, but the error bars are
overlapping. Both profiles
reached their tmax, the time point of the highest concentration, after 18 h.
Clearance of FP1 was faster
which results in congruency at 72 h.
The discrepancy of the ELISA profiles and the different N-glycosylation was
investigating for a possible
relationship.
Comparison of individual N-glycan PK profiles with average ELISA data
Relative concentration of ELISAs was determined by normalizing average
profiles to the maximal
concentration. The relative concentration of the individual N-glycan profiles
was determined by
normalizing the average profiles, based on the determined L/H ratios, to the
maximal L/H ratio. Figure 8
shows the average PK profiles of the major N-glycan of the Fc part (Figure 8A)
and the receptor part
(Figure 8B) of both fusion proteins FP1 and FP2 determined by nanoLC-MS
(filled squares) compared
to the appropriate average ELISA profiles (filled diamonds).
The variability of data of different animals was high which is reflected by
the error bars, with a higher
variability for the N-glycans located at the receptor part. N-glycan profiles
of FP1 GOF, FP1 SG2F and
FP2 G1F showed a shift in tmax towards later time points compared to the
respective ELISA. Only FP2
SG2F had an identical tmax. This course indicates a prolonged serum half-life
with a decreased
clearance rate. However, glycan PK profiles of FP2 were closer to the ELISA
profile than the glycan PK
profiles of FP1.
Glycan maps of FP1 and FP2
The glycan maps were obtained on the basis of the L/H ratios determined by
nanoLC-MS. With the
known distribution of the spiked stable heavy isotope 2-AA labeled N-glycan
standard the glycan maps
41
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
were calculated with high precision. The glycan maps of the Fc and receptor
part from FP1 are shown
in Figure 9. The N-glycans of the Fc part were mainly neutral complex
biantennary N-glycans. Major
glycans were GOF (open diamond) and G1F (filled triangle). All other N-glycans
were below 10%. The
percentages of the glycans of the FC part were constant over time.
The receptor part contains higher portions of acidic N-glycans. SG2F (open
square) was the most
abundant N-glycan. In contrast to the Fc part, the percentages of the receptor
part changed over time.
The portions of GOF (open diamond) and G2F (open triangle) decreased over
time, whereas the
portions of SG2F (open square) and SG2 (filled diamond) showed a trend towards
higher percentages.
The deviation at 32h (Fc part) was most likely due to variations during the
affinity purification or
nanoLC-MS analysis and no temporary change of the N-glycosylation pattern.
Similar results were obtained for FP2. The resulting glycan maps were shown in
Figure 10. For the Fc
part the glycan map show that the major N-glycans G1F (filled triangle), G2F
(open triangle), GOF (open
diamond) and SG2F (open square) and all minor abundant glycans had constant
percentages. The
receptor part showed a different picture. The major N-glycan SG2F (open
square) increased slightly
(relative to other glycans) and the N-glycan G2F (open triangle) decreased
drastically over time (relative
to other glycans).
Shifts of minor abundant N-glycans are discussed in the following. The
variation observed between 6h
and 12h at the Fc part was most likely due to the low protein concentration
resulting in a higher
variation and not due to a significant change of the N-glycan percentages.
Terminal galactosylation of the receptor part accelerated clearance
N-glycans with terminal galactosylation (G1F and G2F) and the N-glycan GOF
carry a terminal N-
acetylglucosamine. G2F showed decreasing percentages over time for FP1 and FP2
(Figures 9 and 10,
right). G1F decreased slightly when localized on FP1, but stayed constant when
bound to FP2 (Figures
9 and 10, right). The same observation was made for GOF (Figures 9 and 10,
right). These results
clearly indicate that FP1 and FP2 with terminal galactosylated G2F were
specifically removed from
circulation. The mechanism behind this clearance could be the
asialoglycoprotein receptor. This
receptor binds selectively to exposed galactose residues. The receptor
furthermore exhibits decreasing
affinity from high to small number of terminal galactose residues, which could
explain the observations
for the less pronounced decrease or constant portion of G1F respectively.
The difference between FP1 and FP2 for the G1F and GOF N-glycan may occur due
to different N-
glycosylation site occupancy on the protein and a different accessibility for
receptors to bind.
Capping N-glycans with sialic acids at the receptor part increased half-life
As described earlier the major N-glycan SG2F had a slight trend towards higher
relative percentages.
At FP1 the amount of 5G2 increased over time and at FP2 and SG1F showed the
same trend. The
other acidic N-glycan had constant percentages (Figures 9 and 10, right).
42
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
These observations lead to the conclusion that the addition of at least one
terminal sialic acid to the
galactosylated N-glycan structure may prevent the protein from selective
clearance. The discrepancy
between FP1 and FP2 with respect to the different N-glycans that increased
over time may again occur
due to differences in the glycosylation site where the structures are
attached.
Sialylation, galactosylation and terminal GIcNAc of FP1 and FP2
An overview over the terminal groups of all attached N-glycans is given in
Figure 11. Minor abundant N-
glycans were also included. The N-glycans were divided into three different
groups: The group
sialylation contains all N-glycans with one or two terminal sialic acids, the
group galactosylation
contains all N-glycans with at least one terminal galactose moiety and the
group terminal GIcNAc
contains N-glycans with a terminal N-acetylglucosamine. For N-glycans with two
characteristics it had to
be decided which attribute is more likely to have an influence on PK, e.g. for
SG1F which has a terminal
sialic acid on one arm and a terminal GIcNAc on the other arm it was decided
that it is part of the
sialylation group because it is more likely that this N-glycan behaves more
like a sialic glycan than like a
terminal GIcNAc glycan due to the shielding effect of the attached terminal
sialic acid group. The same
applies for G1F where the terminal galactose residue has a higher impact than
the terminal GIcNAc.
Terminal groups did not change at the Fc parts of FP1 and FP2. The two
deviations at 6 h (FP2) or at
32 h (FP1) were most probably due to method variation at low target protein
concentrations. In contrast
to the Fc part, the terminal groups at the receptor part showed decreasing
galactosylation which was
compensated with increasing terminal sialylation. N-glycans belonging to the
GIcNAc group were
constant.
Conclusion
A work-flow has been established for the recovery of a fusion protein from
serum samples. In a first
affinity step using Protein G chromatography the complexity of the serum was
drastically reduced.
Serum related proteins that tend to bind non-specific to the sepharose resin
were removed as well as all
non IgG Fc part containing proteins. The pre-cleared serum was applied to
antigen columns and after
washing and enzymatic elution of the Fc part, the N-glycans of the two protein
parts were analyzed
individually. Due to the incorporation of stable heavy isotope 2-AA labeled
glycan standards, variations
during N-glycan processing and nanoLCMS analysis were compensated, and the
time course of
individual glycans in the study were determined with high precision.
Two different batches of a fusion protein were compared in a rabbit study and
the PK profiles of each N-
glycan and the relative distribution of the N-glycans over time were
determined. The results show that
fusion proteins with terminal galactosylated N-glycans that are located at the
receptor part are cleared
faster than the average molecule and that terminal sialic acids increase the
serum half-life. This
clearance is most likely a receptor mediated process of the asialoglycprotein
receptor, which is located
in the liver and known for its affinity to terminal galactosylated N-glycans.
43
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
In summary, this study showed that N-glycosylation of fusion proteins
influences the PK. Terminal
galactosylated N-glycans are cleared faster whereas terminal sialylation
increases serum half-life. This
is most likely due to asialo-glycoprotein receptor mediated binding and occurs
only if the N-glycans are
located at the receptor part. Fc part N-glycosylation (complex bi-antennary)
has no influence on PK
which is in agreement with previous studies. Thus, this method allows detailed
site specific N-glycan PK
profiling of biopharmaceuticals. It was further demonstrated that this kind of
investigation is
advantageous in contrast to N-glycan profiling without site or domain
specificity.
In view of the above, it will be appreciated that the present invention also
relates to the following items:
Items
1. A method of analyzing glycans of a recombinant glycoprotein of interest in
liquid samples of a
mammal comprising
a) providing two or more liquid samples of a mammal comprising the recombinant
glycoprotein of
interest;
b) immobilizing the recombinant glycoprotein of each of said samples on a
separate solid support
coupled to an affinity ligand specific for the recombinant glycoprotein in the
samples;
c) releasing a glycan containing fragment of the recombinant glycoprotein of
each of said samples
from the solid support into separate eluates by enzymatic cleavage of the
recombinant
glycoprotein;
d) optionally releasing the glycans from the glycan containing fragment of the
recombinant
glycoprotein in the separate eluates;
e) labeling the glycans of each of said samples with a first stable isotope of
a fluorescent label; and
f) analyzing the labeled glycans of step e) of each of said samples separately
using LC-MS and
comparing said two or more samples,
wherein a reference standard comprising glycans labeled with a second stable
isotope of the
fluorescent label is added to each of said samples prior to step c), d), e) or
f) and the reference
standard is analyzed together with the labeled glycans of step f).
2. The method of item 1, further comprising analyzing the glycans of a
second glycan containing
fragment of the recombinant glycoprotein of each of said samples that remains
immobilized to the
solid support in step c), comprising the following steps:
aa) pre-clearing each of the two or more liquid samples of a mammal to remove
glycoproteins
binding non-specifically to the solid support prior to step b);
dd) releasing the glycans from the immobilized second glycan containing
fragment of the
recombinant glycoprotein of each of said samples remaining on the solid
support following step c)
into separate solutions;
44
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
ee) labeling the glycans of each of said samples with a first stable isotope
of a fluorescent label;
and
if) analyzing the labeled glycans of step ee) of each of said samples
separately using LC-MS and
comparing said two or more samples,
wherein a reference standard comprising glycans labeled with a second stable
isotope of the
fluorescent label is added to each of said samples prior to step dd), ee) or
if) and wherein the
reference standard is analyzed together with the labeled glycans of step if).
3. The method of item 2, wherein the pre-clearing step aa) comprises
contacting the samples with a
solid support, wherein the solid support is made of the same material as the
solid support used in
step b) of item 1, but is not coupled to said affinity ligand.
4. The method of any one of items 1 to 3, wherein the glycans to be
analyzed are N-glycans selected
from the group consisting of high mannose type, hybrid type or complex type N-
glycans.
5. The method of any one of the preceding items, wherein the fluorescent label
has a stable light and
a stable heavy isotope variant and preferably the reference standard comprises
glycans labeled
with the stably heavy isotope of the fluorescent label.
6. The method of any one of the preceding items, wherein the fluorescent
label is a 12[C] and 13[C]
isotopic pair, preferably a 12[C6] and 13[C6] isotopic pair, more preferably a
12[C6]-2-aminobenzoic
acid (12[C6]-2-AA) and a13[C6]-2-aminobenzoic acid (13[C6]-2-AA) isotopic
pair.
7. The method of any one of the preceding items, wherein the reference
standard comprises at least
the same glycan structures as analyzed in the sample.
8. The method of any one of the preceding items, wherein the method analyzes
glycan structures
individually.
9. The method of any one of the preceding items, wherein the recombinant
glycoprotein is a fusion
protein or an antibody.
10. The method of item 9, wherein the recombinant glycoprotein is an Fc-fusion
protein or an antibody.
11. The method of any one of the preceding items, wherein the glycan
containing fragment of the
recombinant glycoprotein released from the solid support in step c) is an Fc
domain.
12. The method of item 10 or 11, wherein the recombinant glycoprotein is an
antibody and the variable
region of the antibody binds to the affinity ligand immobilized on the solid
support.
13. The method of item 12, wherein the affinity ligand is an antigen.
14. The method of any one of items 10 to 13, wherein the recombinant
glycoprotein is an antibody,
preferably an IgG antibody, preferably an IgG1 or IgG2 antibody, more
preferably a human or
humanized IgG1 or IgG2 antibody.
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
15. The method of items 10 or 11, wherein the recombinant glycoprotein is an
Fc-fusion protein
comprising an Fc-domain and an effector domain and the effector domain binds
to the affinity ligand
immobilized on the solid support, wherein the affinity ligand is a binding
partner or an antibody
specifically binding to the effector domain of the Fc-fusion protein.
16. The method of any one of items 10 to 15, wherein the enzyme used to
release the glycan
containing fragment of the recombinant glycoprotein from the solid support is
an endoproteinase,
preferably papain, ficin, cysteine protease SpeB (FabULOUS) or cysteine
proteinase IdeS
(FabRICATORO), more preferably cysteine proteinase IdeS (FabRICATORO).
17. The method of any one of items 10 to 15, wherein the enzyme used to
release the glycan
containing fragment of the recombinant glycoprotein from the solid support is
a glycanase,
preferably EndoSGlycanase (IgGZER0).
18. The method of any one of items 10 to 17, wherein the pre-clearing step
comprises
aai) immobilizing Fc-containing glycoproteins on a solid support using an Fc-
binding protein,
wherein said Fc-binding protein is preferably selected from protein G or
protein A, more preferably
said Fc-binding protein is protein G; and
aaii) eluting the Fc-containing glycoproteins;
wherein the solid support used in said pre-clearing step is made of the same
material as the solid
support used in step b) of item 1, but is not coupled to said affinity ligand.
19. The method of any one of the preceding items, wherein the LC-MS is a
reverse phase LC-MS.
20. The method of any one of the preceding items, wherein the LC-MS is a
NanoLC-MS, preferably a
reverse phase NanoLC-MS.
21. The method of any one of the preceding items, wherein the solid support is
a resin comprising
microbeads, preferably sepharose beads, agarose beads or magnetic beads, more
preferably
sepharose beads.
22. The method of any one of the preceding items, wherein the affinity ligand
is coupled to the solid
support via N-hydroxysuccinimide (NHS), cyanogen bromide, epoxy, carbodiimide
or thiopropyl,
preferably via N-hydroxysuccinimide (NHS).
23. The method of any one of the preceding items, wherein the two or more
liquid samples of a
mammal are prepared for analysis in a multi well filter plate, preferably in a
24, 96 or 384 well filter
plate, more preferably in a 96 well filter plate.
24. The method of any one of the preceding items, wherein the liquid samples
of a mammal are body
fluids, preferably selected from the group consisting of serum or plasma,
urine, cerebral spinal fluid,
amniotic fluid, saliva, sweat, ejaculate, tears, phlegm, vaginal secretion,
vaginal wash and colonic
wash; preferably wherein said samples are plasma or serum samples.
46
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
25. The method of any one of the preceding items, wherein the two or more
liquid samples of a
mammal were obtained from the same subject.
26. The method of item 25, wherein pharmacokinetic parameters of at least one
specific glycan
structure of the glycans of the recombinant glycoprotein of interest are
determined, preferably the
Cmax, tmax, AUG or 1112.
27. The method of items 25 or 26, wherein the two or more liquid samples of a
mammal were obtained
at different time points from the same subject.
28. The method of item 27, wherein at least about 24 h, 48 h, 72 h, 96 h, 120
h, 144 h, 168 h, 336 h,
504 h, 672 h, 840 h are between the first and the last time point.
29. The method of any one of the preceding items, wherein the two or more
liquid samples of a
mammal are 5 or more, 10 or more, 20 or more, 30 or more, 40 or more or 50 or
more samples.
30. The method of any one of the preceding items, wherein
(i) the area under the curve of the extracted ion chromatograms (EICs) of the
glycans from each of
said samples labeled with the first stable isotope of the fluorescent label
and its respective
reference standard glycan labeled with the second isotope of the fluorescently
label is determined
to form a ratio of first isotope sample glycan to second isotope reference
standard over time, and
(ii) the N-glycan percentages based on the known relative distribution of the
individual glycans in the
reference standard are calculated for each of said samples.
31. The method of any one of the preceding items, wherein the glycans are
analyzed in an aliquot of
each of said two or more liquid samples of a mammal.
32. The method of any one of the preceding items, wherein the samples or the
aliquots to be analyzed
have a volume from about 1 pl to about 1000 pl, from about 5 pl to about 500
pl, from about 10 pl to
about 200 pl, from about 10 pl to about 100 pl, from about 25 pl to about 100
pl, or from about 40 pl
to about 75 pi, preferably of about 50 pl.
33. The method of item 31 or 32, wherein the concentration of said recombinant
glycoprotein is
analyzed in a further aliquot of said two or more liquid samples of a mammal.
34. The method of item 33, wherein the concentration of said recombinant
glycoprotein is analyzed by
ELISA.
35. The method of any one of the preceding items, wherein the mammalian liquid
sample is a human, a
monkey, a rodent, a dog, a cat or a pig sample.
36. The method of item 35, wherein the mammalian liquid sample is a rodent
sample and the rodent is
a mouse, a rat, a hamster or a rabbit.
47
CA 02954364 2017-01-05
WO 2016/009077 PCT/EP2015/066488
37. The method of any one of the preceding items, wherein the method allows
for analyzing attomolar
concentrations of the individual labeled glycans, e.g., concentrations as low
as 800 amol, preferably
as low as 600 amol, and more preferably as low as 400 amol.
38. The method of any one of the preceding items, wherein the method allows
for analyzing the glycans
of the recombinant glycoprotein at a recombinant glycoprotein concentration in
each of the two or
more liquid samples of a mammal of 10 pg/ml or less, 5 pg/ml or less, 2 pg/ml
or less, 1 pg/ml or
less, 0.5 pg/ml or less, 0.2 pg/ml or less or 0.1 pg/ml or less.
39. The method of any one of the preceding items, wherein the method allows
for analyzing the glycans
of the recombinant glycoprotein in the two or more liquid samples of a mammal
comprising 0.5 pg
or less, 0.25 pg or less, 0.1 pg or less. 0.05 pg or less, 0.025 pg or less,
0.01 pg or less, or 0.005
pg or less of the recombinant protein each.
40. The method of any one of the preceding items, wherein at least steps b),
c), d) and e) and/or at
least steps aa), dd) and ee) are operated in a high throughput manner.
41. A method of preparing a glycoprotein based pharmaceutical composition
comprising analyzing the
glycans of a recombinant glycoprotein in liquid samples of a mammal according
to any one of the
preceding items, and formulating the glycoprotein into said pharmaceutical
composition.
42. A method of analyzing glycans of an Fc-fusion protein of interest in
liquid samples of a mammal,
comprising
a) providing two or more liquid samples from a mammal comprising an Fc-fusion
protein containing
an Fc-domain and an effector domain;
aa) pre-clearing each of the two or more liquid samples of a mammal
comprising:
i) immobilizing the Fc-fusion protein of each of said samples on a separate
solid support using
an Fc-binding protein, wherein said Fc-binding protein is preferably selected
from protein G or
protein A, more preferably said Fc-binding protein is protein G; and
ii) eluting the Fc-fusion protein;
b) immobilizing the Fc-fusion protein of each of said samples on a separate
solid support coupled
to an affinity ligand specific for the effector domain of the Fc-fusion
protein in the sample, wherein
the affinity ligand is a binding partner or an antibody specifically binding
to the effector domain of
the Fc-fusion protein;
c) releasing the Fc-domain of the Fc-fusion protein of each of said samples
from the solid support
into separate eluates by cleaving with an endopeptidase specific for Fc-fusion
proteins;
dd) releasing the glycans from the immobilized effector domain of the Fc-
fusion protein of each of
said samples remaining on the solid support following step c) into separate
solutions;
48
CA 02954364 2017-01-05
WO 2016/009077
PCT/EP2015/066488
ee) labeling the glycans of each of said samples with a first stable isotope
of a fluorescent label;
and
if) analyzing the labeled glycans of step ee) of each of said samples
separately using LC-MS and
comparing said two or more samples;
wherein a reference standard comprising glycans labeled with a second stable
isotope of the
fluorescent label is added to each of said samples prior to step dd), ee) or
if) and wherein the
reference standard is analyzed together with the labeled glycans of step if).
49
